diff --git a/.DS_Store b/.DS_Store
new file mode 100644
index 0000000000000000000000000000000000000000..9a302793c3afc4b45f3ced644c6a9b60b9cf23ae
Binary files /dev/null and b/.DS_Store differ
diff --git a/.ipynb_checkpoints/MORSTS-checkpoint.ipynb b/.ipynb_checkpoints/MORSTS-checkpoint.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..a74128370703d43e2d023bef516017cbb9d140a2
--- /dev/null
+++ b/.ipynb_checkpoints/MORSTS-checkpoint.ipynb
@@ -0,0 +1,3251 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import numpy as np\n",
+ "from sklearn.linear_model import LinearRegression\n",
+ "from sklearn.neighbors import KNeighborsRegressor\n",
+ "from sklearn.ensemble import RandomForestRegressor\n",
+ "from sklearn.tree import DecisionTreeRegressor\n",
+ "import os\n",
+ "from sklearn.metrics import mean_squared_error\n",
+ "import random\n",
+ "import math\n",
+ "import time\n",
+ "# from keras.models import Sequential\n",
+ "# from keras.layers import Dense\n",
+ "# from keras.layers import LSTM\n",
+ "from sklearn.decomposition import PCA\n",
+ "from sklearn.model_selection import KFold\n",
+ "import matplotlib.pyplot as plt\n",
+ "from random import choice\n",
+ "import pandas as pd\n",
+ "from sklearn.linear_model import Lasso\n",
+ "# from msvr import kernelmatrix\n",
+ "# from msvr import msvr"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# class MSVR_F:\n",
+ "# def __init__(self,ker = 'rbf', C = 2,epsi = 0.001,par = 0.8,tol = 1e-10):\n",
+ "# self.ker = ker\n",
+ "# self.C = C\n",
+ "# self.epsi = epsi\n",
+ "# self.par = par\n",
+ "# self.tol = tol\n",
+ " \n",
+ "# def fit(self,X_train,Y_train):\n",
+ "# X_train = np.array(X_train)\n",
+ "# Y_train = np.array(Y_train)\n",
+ "# self.X_train = X_train\n",
+ "# self.Beta = msvr(X_train, Y_train, self.ker, self.C, self.epsi, self.par, self.tol)\n",
+ " \n",
+ "# def predict(self,X_test):\n",
+ "# X_test = np.array(X_test)\n",
+ "# H = kernelmatrix('rbf', X_test, self.X_train, par);\n",
+ "# return np.dot(H, self.Beta)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Ecrire un générateur de blocs pour obtenir différents ensembles de données initiales\n",
+ "# data est la donnée, n est le nombre de parties en lesquelles elle est divisée, i est la première\n",
+ "# 写一个块生成器，用来获得不同的初始数据集\n",
+ "# data 是数据，n是分成几份，i是第几份\n",
+ "\n",
+ "def getChunks(data,n,i):\n",
+ " return data[ int(len(data)*(i)/n)+1 : (int(len(data)*(i+1)/n)-1)]\n",
+ "def getPourCentage(data,beginCentage,endCentage):\n",
+ " return data[ (int(len(data)*beginCentage)): (int(len(data)*endCentage))]\n",
+ "#功能介绍：返回初始数据，返回全部训练数据，判断是否还有下一个batch，返回当前batch，\n",
+ "#Description de la fonction : retour des données initiales, retour de toutes les données d'entraînement, détermination de l'existence d'un lot suivant, retour du lot actuel.\n",
+ "class feeder_Ini_Train_Batch():\n",
+ " def __init__(self,X,Y,beginCentage,endCentage,batch_size):\n",
+ " self.X = X\n",
+ " self.Y = Y\n",
+ " #elf.X_train = getPourCentage(X,0,iniCentage)\n",
+ " #elf.Y_train = getPourCentage(Y,0,iniCentage)\n",
+ " self.X_test = getPourCentage(X,beginCentage,endCentage)\n",
+ " self.Y_test = getPourCentage(Y,beginCentage,endCentage)\n",
+ " self.batch_size = batch_size\n",
+ " self.t = 1 # t C'est l'indexation et le temps. 就是索引和时间\n",
+ " def getIni_X_Y(self,iniCentage):\n",
+ " return getPourCentage(self.X,0,iniCentage),getPourCentage(self.Y,0,iniCentage)\n",
+ " def getTrain_X_Y(self):\n",
+ " return self.X_test,self.Y_test\n",
+ " def hasThisBatch(self):\n",
+ " if (self.t)*self.batch_size < len (self.X_test):\n",
+ " return True\n",
+ " else:\n",
+ " return False\n",
+ " def hasThisBatch_and_nextBath(self):\n",
+ " if (self.t+1)*(self.batch_size) < len (self.X_test):\n",
+ " return True\n",
+ " else:\n",
+ " return False\n",
+ " def getThisBatch(self):\n",
+ " if self.hasThisBatch() == True:\n",
+ " actul_Batch_X = self.X_test[(self.t-1)*self.batch_size:(self.t)*self.batch_size]\n",
+ " actul_Batch_Y = self.Y_test[(self.t-1)*self.batch_size:(self.t)*self.batch_size]\n",
+ " return actul_Batch_X,actul_Batch_Y\n",
+ " else:\n",
+ " print('err index out')\n",
+ " def getNextBatch_getThisBatch(self):\n",
+ " if self.hasThisBatch() == True:\n",
+ " next_Batch_X = self.X_test[(self.t)*self.batch_size:(self.t+1)*self.batch_size]\n",
+ " next_Batch_Y = self.Y_test[(self.t)*self.batch_size:(self.t+1)*self.batch_size]\n",
+ " return next_Batch_X,next_Batch_Y\n",
+ " else:\n",
+ " print('err index out')\n",
+ " def goNext(self):\n",
+ " self.t +=1\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def getMeanSuperPourCentage(data,PourCentage): # Calculez la valeur moyenne du pourcentage supérieur des données. 计算排名前百分之多少的数据的均值\n",
+ "# print(np.percentile(data, (PourCentage)))\n",
+ " return np.mean([x for x in data if x >= np.percentile(data, (PourCentage))])\n",
+ "def getMeanSuperPourCentage_Martix_ParLine(data,PourCentage):\n",
+ " return np.array([getMeanSuperPourCentage(x,PourCentage) for x in data ])\n",
+ "\n",
+ " # 0子模型 1编号时间 2权重 3class名 4俗称 5varepsilon加权错 6 Omega时间权重 7err普通RMSE错误 8 模型地址\n",
+ " # 0 submodèle 1 temps de numérotation 2 poids 3 nom de classe 4 nom commun 5 erreur de pondération varepsilon 6 pondération temps oméga 7 erreur RMSE commune 8 adresse du modèle\n",
+ "\n",
+ "def RandonSelectionModle(index,x,y , optitionInnErrOrCroissErr,numFlod = 3, indique_sousModle = 'Random_LS_LI_R_T'): \n",
+ "\n",
+ " if indique_sousModle == 'Random_LI_R_T_KNN':\n",
+ " indique_sousModle = choice(['Lasso','R_Forest','Tree','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_R_T_KNN':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','R_Forest','Tree','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_R_T_SVR':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','R_Forest','Tree','SVR'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_R_T':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','R_Forest','Tree'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_R_KNN':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','R_Forest','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_T_KNN':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','Tree','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_R_T_KNN':\n",
+ "# indique_sousModle = choice(['LSTM','R_Forest','Tree','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_R_T':\n",
+ "# indique_sousModle = choice(['LSTM','R_Forest','Tree'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_T':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','Tree'])\n",
+ " \n",
+ " print('Randomly selected indique_sousModle ',indique_sousModle)\n",
+ " weight = 1 # weight of this base model 投票权重\n",
+ " varepsilon = 1 # Résultats des fonctions de perte 丢失函数结果 也就是 varepsilon 加权之后的错\n",
+ " omega = 1 # la pondération du temps 关于时间的权重\n",
+ " RMSE = 1 # err 普通RMSE错误\n",
+ " x_rnn = np.reshape(x, (x.shape[0], 1,x.shape[1]))\n",
+ " kf = KFold(n_splits=numFlod)\n",
+ " yhat_all_flod = []\n",
+ " \n",
+ "# if indique_sousModle == 'LSTM' :\n",
+ "# modele = Sequential()\n",
+ "# modele.add(LSTM(500, input_shape =(1, x_rnn.shape[2]) , activation='relu'))\n",
+ "# modele.add(Dense(500, activation='relu'))\n",
+ "# modele.add(Dense(y.shape[1] , activation='sigmoid'))\n",
+ "# # Compile model\n",
+ "# modele.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
+ "# modele.fit(x_rnn, y, epochs=6, batch_size=2000, verbose=2)\n",
+ "# if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ "# for train, test in kf.split(x_rnn):\n",
+ "# X_train_numFlod = np.array( [x_rnn[i] for i in train])\n",
+ "# Y_train_numFlod = np.array([y[i] for i in train])\n",
+ "# X_test_numFlod = np.array([x_rnn[i] for i in test])\n",
+ "# Y_test_numFlod = np.array([y[i] for i in test])\n",
+ "# modele_numFlod = Sequential()\n",
+ "# modele_numFlod.add(LSTM(500, input_shape =(1, x_rnn.shape[2]) , activation='relu'))\n",
+ "# modele_numFlod.add(Dense(500, activation='relu'))\n",
+ "# modele_numFlod.add(Dense(y.shape[1] , activation='sigmoid'))\n",
+ "# modele_numFlod.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
+ "# modele_numFlod.fit(X_train_numFlod, Y_train_numFlod, epochs=4, batch_size=1000, verbose=2)\n",
+ "# yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ "# yhat_all_flod += yhat__numFlod.tolist()\n",
+ "# return [modele,index,weight,modele.__class__,'LSTM',varepsilon,omega,RMSE,id(modele)] , yhat_all_flod\n",
+ " \n",
+ " \n",
+ " \n",
+ "# if indique_sousModle == 'SVR' :\n",
+ "# modele = MSVR_F()\n",
+ "# modele.fit(x,y)\n",
+ "# if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ "# for train, test in kf.split(x):\n",
+ "# X_train_numFlod = np.array( [x[i] for i in train])\n",
+ "# Y_train_numFlod = np.array([y[i] for i in train])\n",
+ "# X_test_numFlod = np.array([x[i] for i in test])\n",
+ "# Y_test_numFlod = np.array([y[i] for i in test])\n",
+ "# modele_numFlod = MSVR_F()\n",
+ "# modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ "# yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ "# yhat_all_flod += yhat__numFlod.tolist() \n",
+ "# return [modele,index,weight,modele.__class__,'SVR',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ "\n",
+ " elif indique_sousModle == 'KNN':\n",
+ " modele = KNeighborsRegressor(n_neighbors=15)\n",
+ " modele.fit(x,y)\n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " for train, test in kf.split(x):\n",
+ " X_train_numFlod = np.array( [x[i] for i in train])\n",
+ " Y_train_numFlod = np.array([y[i] for i in train])\n",
+ " X_test_numFlod = np.array([x[i] for i in test])\n",
+ " Y_test_numFlod = np.array([y[i] for i in test])\n",
+ " modele_numFlod = KNeighborsRegressor(n_neighbors=15)\n",
+ " modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ " yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ " yhat_all_flod += yhat__numFlod.tolist() \n",
+ " return [modele,index,weight,modele.__class__,'KNeighborsRegressor',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ " elif indique_sousModle == 'R_Forest':\n",
+ " modele = RandomForestRegressor(n_estimators=10, )\n",
+ " modele.fit(x,y)\n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " for train, test in kf.split(x):\n",
+ " X_train_numFlod = np.array( [x[i] for i in train])\n",
+ " Y_train_numFlod = np.array([y[i] for i in train])\n",
+ " X_test_numFlod = np.array([x[i] for i in test])\n",
+ " Y_test_numFlod = np.array([y[i] for i in test])\n",
+ " modele_numFlod = RandomForestRegressor(n_estimators=10,)\n",
+ " modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ " yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ " yhat_all_flod += yhat__numFlod.tolist()\n",
+ " return [modele,index,weight,modele.__class__,'RandomForestRegressor',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ " elif indique_sousModle == 'Lasso':\n",
+ " modele = Lasso(alpha=0.1)\n",
+ " modele.fit(x,y)\n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " for train, test in kf.split(x):\n",
+ " X_train_numFlod = np.array( [x[i] for i in train])\n",
+ " Y_train_numFlod = np.array([y[i] for i in train])\n",
+ " X_test_numFlod = np.array([x[i] for i in test])\n",
+ " Y_test_numFlod = np.array([y[i] for i in test])\n",
+ " modele_numFlod = Lasso(alpha=0.1)\n",
+ " modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ " yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ " yhat_all_flod += yhat__numFlod.tolist() \n",
+ " return [modele,index,weight,modele.__class__,'LassoRegression',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ " elif indique_sousModle == 'Tree':\n",
+ " modele = DecisionTreeRegressor()\n",
+ " modele.fit(x,y)\n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " for train, test in kf.split(x):\n",
+ " X_train_numFlod = np.array( [x[i] for i in train])\n",
+ " Y_train_numFlod = np.array([y[i] for i in train])\n",
+ " X_test_numFlod = np.array([x[i] for i in test])\n",
+ " Y_test_numFlod = np.array([y[i] for i in test])\n",
+ " modele_numFlod = DecisionTreeRegressor()\n",
+ " modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ " yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ " yhat_all_flod += yhat__numFlod.tolist() \n",
+ " return [modele,index,weight,modele.__class__,'DecisionTreeRegressor',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ "def makePredictionModele(modele,x):\n",
+ " if type(modele) in [type(RandomForestRegressor()), \n",
+ " type(KNeighborsRegressor()),\n",
+ " type(Lasso()),\n",
+ " type(DecisionTreeRegressor()),\n",
+ " type(MSVR_F())] :\n",
+ " return modele.predict(x)\n",
+ " else:\n",
+ " print('报错，不在列表中 Error reported, not in the list')\n",
+ "\n",
+ "def beta_fonction(k,t,a,b): \n",
+ " return 1/(1+ math.exp(-a*(t-k-b))) \n",
+ "\n",
+ "def vote_mean(dic_expert, actul_Batch_X, actul_Batch_Y, afficher_detail = False):\n",
+ " H_res = np.array( [[0.0]* len(actul_Batch_Y[0])] *len(actul_Batch_Y))\n",
+ " print(H_res)\n",
+ " sumWeigt = 0\n",
+ " \n",
+ " list_RMSE_SousModele = []\n",
+ "\n",
+ " for key,value in dic_expert.items():\n",
+ " yhat_sousM = makePredictionModele(value[0],actul_Batch_X)\n",
+ " \n",
+ " if np.array(yhat_sousM).ndim == 1:\n",
+ " print('type([[z] for z in yhat_sousM]) ',type([[z] for z in yhat_sousM]))\n",
+ " yhat_sousModel = np.array([[z] for z in yhat_sousM])\n",
+ " else:\n",
+ " yhat_sousModel = yhat_sousM\n",
+ " print(type(yhat_sousModel))\n",
+ " print('Output of each sub-model',yhat_sousModel)\n",
+ " H_res =H_res + yhat_sousModel*value[2]\n",
+ " sumWeigt += value[2] # 2 représente les poids des votes des sous-modèles 2号代表子模型投票权\n",
+ " list_RMSE_SousModele.append(aRMSE( actul_Batch_Y ,yhat_sousModel))\n",
+ " yhat_H = np.array(H_res)/sumWeigt # L'obtention des résultats prédits pour H\n",
+ " \n",
+ "# print('投票结束之后的输出', yhat_H)\n",
+ "# print('Y的原始值', actul_Batch_Y)\n",
+ " if afficher_detail == True:\n",
+ " print( 'list_RMSE_SousModele ', list_RMSE_SousModele, 'mean of list_RMSE_SousModele', np.mean(list_RMSE_SousModele) )\n",
+ " print(' RMSE resultat vote', aRMSE( actul_Batch_Y ,yhat_H))\n",
+ " return yhat_H\n",
+ "\n",
+ "\n",
+ "def vote_OnlyMaxERRWeight(dic_expert, actul_Batch_X, actul_Batch_Y):\n",
+ " the_key = 0\n",
+ " maxWeight = 0\n",
+ " for key,value in dic_expert.items():\n",
+ " if dic_expert[key][5] > maxWeight:\n",
+ " maxWeight = dic_expert[key][5]\n",
+ " the_key = key\n",
+ " yhat_H = makePredictionModele(dic_expert[the_key][0],actul_Batch_X)\n",
+ " err_H = aRMSE(actul_Batch_Y,yhat_H)# err as RMSE \n",
+ " print('Sub-model selected to vote, no. ', the_key, ' 其 RMSE', err_H)\n",
+ " return yhat_H\n",
+ "\n",
+ "def vote_OnlyMaxESpWeight(dic_expert, actul_Batch_X, actul_Batch_Y):\n",
+ " the_key = 0\n",
+ " maxWeight = 0\n",
+ " for key,value in dic_expert.items():\n",
+ " if dic_expert[key][2] > maxWeight:\n",
+ " maxWeight = dic_expert[key][2]\n",
+ " the_key = key\n",
+ " yhat_H = makePredictionModele(dic_expert[the_key][0],actul_Batch_X)\n",
+ " err_H = aRMSE(actul_Batch_Y,yhat_H)# err as RMSE \n",
+ " print('Sub-model selected to vote, no. ', the_key, ' with its RMSE', err_H)\n",
+ " return yhat_H\n",
+ "\n",
+ "def updatingALLSousModele(dic_expert, x, y):\n",
+ " print (' Update sub-model program start' )\n",
+ " x_rnn = np.reshape(x, (x.shape[0], 1,x.shape[1]))\n",
+ " for key,value in dic_expert.items():\n",
+ " if type(value[0]) in [type(Sequential())]:\n",
+ " value[0].fit(x_rnn, actul_Batch_Y)\n",
+ "\n",
+ " \n",
+ "def get_filename(path,filetype): # 输入路径、文件类型例如'.csv'\n",
+ " name = []\n",
+ " for root,dirs,files in os.walk(path):\n",
+ " for i in files:\n",
+ " if os.path.splitext(i)[1]==filetype:\n",
+ " name.append(i) \n",
+ " return name\n",
+ "\n",
+ "def aRMSE(y_true,y_pred):\n",
+ " return np.mean(mean_squared_error(y_true, y_pred, squared=False, multioutput='raw_values'))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Randomly selected indique_sousModle KNN\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00920465 0.01133741 ... 0.01762998 0. 0. ]\n",
+ " [0. 0.00920465 0.01167685 ... 0.01762998 0. 0. ]\n",
+ " [0. 0.00661305 0.01398506 ... 0.01762998 0. 0. ]\n",
+ " ...\n",
+ " [0.49044212 0.37342866 0.32790224 ... 0.03077329 0.12271824 0.47038019]\n",
+ " [0.5155262 0.37229669 0.3233537 ... 0.02866192 0.11845292 0.48364279]\n",
+ " [0.49164068 0.37742032 0.32416836 ... 0.03283188 0.11497078 0.46560566]]\n",
+ "list_RMSE_SousModele [0.08582627654661987] mean of list_RMSE_SousModele 0.08582627654661987\n",
+ " RMSE resultat vote 0.08582627654661987\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 0 value[6] 0.5474158342212556\n",
+ "key 1 value[6] 1.0\n",
+ "key 0 value[2] 0.11835014748206642\n",
+ "key 1 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.08582627654661987\n",
+ "Current Error err_H 0.08582627654661987\n",
+ "myFeeder.t 1\n",
+ "[0.08582627654661987]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.4879594 0.37610962 0.32708758 ... 0.03204012 0.12068197 0.45234306]\n",
+ " [0.39842231 0.37527554 0.31846572 ... 0.03098443 0.13186337 0.44986737]\n",
+ " [0.41828411 0.37569258 0.31758316 ... 0.03056215 0.13094765 0.44756852]\n",
+ " ...\n",
+ " [0. 0.00944296 0.01099796 ... 0.01826339 0. 0. ]\n",
+ " [0. 0.00944296 0.01099796 ... 0.01826339 0. 0. ]\n",
+ " [0. 0.00944296 0.01099796 ... 0.01826339 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.79245398e-01 3.60321716e-01 3.36456212e-01 ... 3.10899974e-02\n",
+ " 1.61829026e-01 5.12643678e-01]\n",
+ " [5.86828105e-01 3.65534704e-01 3.40868975e-01 ... 3.53127474e-02\n",
+ " 1.63961684e-01 5.20601238e-01]\n",
+ " [5.87146089e-01 3.65921954e-01 3.41208418e-01 ... 3.53655318e-02\n",
+ " 1.64046027e-01 5.20954907e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 2.82416836e-02 ... 1.87384534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.68160217e-02 ... 1.86328847e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.25390360e-02 ... 1.85801003e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.08948236022221946, 0.05665463313893739] mean of list_RMSE_SousModele 0.07306849668057842\n",
+ " RMSE resultat vote 0.05673817506635284\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 0 value[6] 0.39107838594488675\n",
+ "key 1 value[6] 0.5474158342212556\n",
+ "key 2 value[6] 1.0\n",
+ "key 0 value[2] 0.0784792037307775\n",
+ "key 1 value[2] 0.4215730539577539\n",
+ "key 2 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07128222580648635\n",
+ "Current Error err_H 0.05673817506635284\n",
+ "myFeeder.t 2\n",
+ "[0.08582627654661987, 0.05673817506635284]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00944296 0.01099796 ... 0.01826339 0. 0. ]\n",
+ " [0. 0.00944296 0.01099796 ... 0.01826339 0. 0. ]\n",
+ " [0. 0.00941317 0.01099796 ... 0.01826339 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00944296 0.01235574 ... 0.01884402 0. 0. ]\n",
+ " [0. 0.00938338 0.01228785 ... 0.0192663 0. 0. ]\n",
+ " [0. 0.00944296 0.01235574 ... 0.01884402 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02111337 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00047662 0.0245757 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00053619 0.02939579 ... 0.01879124 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.01900883 ... 0.02179995 0. 0. ]\n",
+ " [0. 0.00047662 0.01832994 ... 0.02211665 0. 0. ]\n",
+ " [0. 0.00044683 0.01900883 ... 0.02185273 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.02545825 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00049151 0.02556008 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00058088 0.02627291 ... 0.01947743 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00098302 0.0194501 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00116175 0.01965377 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00098302 0.0194501 ... 0.01821061 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.09622021793321976, 0.06015379098027394, 0.07893986467740179] mean of list_RMSE_SousModele 0.07843795786363184\n",
+ " RMSE resultat vote 0.06456156236941377\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 0 value[6] 0.3081780004228992\n",
+ "key 1 value[6] 0.39107838594488675\n",
+ "key 2 value[6] 0.5474158342212556\n",
+ "key 3 value[6] 1.0\n",
+ "key 0 value[2] 0.06408903017092009\n",
+ "key 1 value[2] 0.25575178930892106\n",
+ "key 2 value[2] 0.5114365308966105\n",
+ "key 3 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06904200466079549\n",
+ "Current Error err_H 0.06456156236941377\n",
+ "myFeeder.t 3\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00947274 0.01242363 ... 0.01863288 0. 0. ]\n",
+ " [0. 0.00947274 0.01242363 ... 0.01842175 0. 0. ]\n",
+ " [0. 0.00947274 0.01242363 ... 0.01842175 0. 0. ]\n",
+ " ...\n",
+ " [0.52658228 0.33976765 0.31167685 ... 0.03024545 0.13229713 0.42953139]\n",
+ " [0.52658228 0.33976765 0.31167685 ... 0.03024545 0.13229713 0.42953139]\n",
+ " [0.52600746 0.34432529 0.31765105 ... 0.03066772 0.13457437 0.43678161]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.05023761e-02 ... 2.19055160e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.12491514e-02 ... 2.20638691e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.04344874e-02 ... 2.20110847e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.57701951e-01 3.28060769e-01 3.19076714e-01 ... 2.93481130e-02\n",
+ " 1.94047834e-01 4.41909814e-01]\n",
+ " [6.59940072e-01 3.28388442e-01 3.18397828e-01 ... 2.94008973e-02\n",
+ " 1.93734562e-01 4.44385500e-01]\n",
+ " [6.39907051e-01 3.25141495e-01 3.09572301e-01 ... 2.88730536e-02\n",
+ " 1.93734562e-01 4.40671972e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00075961 0.01965377 ... 0.0178939 0. 0. ]\n",
+ " [0. 0.00075961 0.0203666 ... 0.01749802 0. 0. ]\n",
+ " [0. 0.00084897 0.01985743 ... 0.0178939 0. 0. ]\n",
+ " ...\n",
+ " [0.71216291 0.34016979 0.29613035 ... 0.03159145 0.19701789 0.49071618]\n",
+ " [0.73032471 0.33391421 0.28543788 ... 0.03151227 0.19790349 0.48514589]\n",
+ " [0.7083104 0.34374441 0.30142566 ... 0.03182898 0.19739743 0.49151194]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.63747454e-02 ... 2.00316706e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.66802444e-02 ... 1.92399050e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.64765784e-02 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.38525041e-01 3.41510277e-01 3.08044807e-01 ... 3.10372130e-02\n",
+ " 1.59081872e-01 4.71883289e-01]\n",
+ " [6.35699872e-01 3.42672029e-01 3.02443992e-01 ... 3.08788599e-02\n",
+ " 1.58901139e-01 4.74535809e-01]\n",
+ " [6.09594570e-01 3.41912422e-01 3.13849287e-01 ... 3.08788599e-02\n",
+ " 1.59154166e-01 4.79575597e-01]]\n",
+ "list_RMSE_SousModele [0.11489148885061383, 0.07376921430269853, 0.08692241043370037, 0.06255720507186519] mean of list_RMSE_SousModele 0.08453507966471947\n",
+ " RMSE resultat vote 0.06281071100358247\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "key 0 value[6] 0.25465259711919624\n",
+ "key 1 value[6] 0.3081780004228992\n",
+ "key 2 value[6] 0.39107838594488675\n",
+ "key 3 value[6] 0.5474158342212556\n",
+ "key 4 value[6] 1.0\n",
+ "key 0 value[2] 0.058368766099088025\n",
+ "key 1 value[2] 0.19187708503568446\n",
+ "key 2 value[2] 0.3541199055891011\n",
+ "key 3 value[2] 0.5941616353358824\n",
+ "key 4 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06748418124649223\n",
+ "Current Error err_H 0.06281071100358247\n",
+ "myFeeder.t 4\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.52893047 0.34614239 0.3173795 ... 0.03066772 0.13679137 0.43872679]\n",
+ " [0.52893047 0.34614239 0.3173795 ... 0.03066772 0.13679137 0.43872679]\n",
+ " [0.52893047 0.34614239 0.3173795 ... 0.03066772 0.13679137 0.43872679]\n",
+ " ...\n",
+ " [0. 0.00935359 0.01113374 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.01018767 0.01072641 ... 0.01842175 0. 0. ]\n",
+ " [0. 0.01018767 0.01072641 ... 0.01842175 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.16951018e-01 3.20226393e-01 2.99185336e-01 ... 2.78701504e-02\n",
+ " 1.93288752e-01 4.43678161e-01]\n",
+ " [6.16951018e-01 3.20226393e-01 2.99185336e-01 ... 2.78701504e-02\n",
+ " 1.93288752e-01 4.43678161e-01]\n",
+ " [6.18296337e-01 3.23056300e-01 2.99049559e-01 ... 2.78701504e-02\n",
+ " 1.92951383e-01 4.50751547e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 2.62050238e-02 ... 1.92135128e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.60013578e-02 ... 1.92135128e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.58655804e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.08732343e-01 3.44280608e-01 3.00712831e-01 ... 3.19081552e-02\n",
+ " 1.97090186e-01 4.92838196e-01]\n",
+ " [7.07980187e-01 3.44459339e-01 3.00814664e-01 ... 3.19081552e-02\n",
+ " 1.96728719e-01 4.93103448e-01]\n",
+ " [7.07154651e-01 3.47676497e-01 3.04378819e-01 ... 3.11955661e-02\n",
+ " 1.96385324e-01 4.92572944e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 6.25558534e-04 2.65784114e-02 ... 1.89231987e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.59674134e-02 ... 1.95566112e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 6.25558534e-04 2.65784114e-02 ... 1.89231987e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.39992662e-01 3.33467382e-01 2.90224033e-01 ... 3.06413302e-02\n",
+ " 1.39851798e-01 4.53580902e-01]\n",
+ " [6.46064942e-01 3.32439678e-01 3.05906314e-01 ... 3.11955661e-02\n",
+ " 1.60075908e-01 4.66578249e-01]\n",
+ " [6.38671803e-01 3.36997319e-01 2.96334012e-01 ... 3.13539192e-02\n",
+ " 1.39580698e-01 4.59416446e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.36193029e-04 2.55600815e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.58655804e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.80875782e-04 2.59674134e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.61462729e-01 3.42686923e-01 3.42158859e-01 ... 3.48376880e-02\n",
+ " 1.99096331e-01 4.38019452e-01]\n",
+ " [7.60594386e-01 3.36103664e-01 3.33401222e-01 ... 3.48904724e-02\n",
+ " 1.98361347e-01 4.38373121e-01]\n",
+ " [7.72127438e-01 3.33899315e-01 3.30957230e-01 ... 3.50488255e-02\n",
+ " 1.98144467e-01 4.36604775e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 1.54107264e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.54107264e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.54786151e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.10226191733611455, 0.09137311316122942, 0.1100676913925561, 0.1000344392177678, 0.09868809559308724] mean of list_RMSE_SousModele 0.10048505134015102\n",
+ " RMSE resultat vote 0.0888887469703822\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 0 value[6] 0.21615527050978905\n",
+ "key 1 value[6] 0.25465259711919624\n",
+ "key 2 value[6] 0.3081780004228992\n",
+ "key 3 value[6] 0.39107838594488675\n",
+ "key 4 value[6] 0.5474158342212556\n",
+ "key 5 value[6] 1.0\n",
+ "key 0 value[2] 0.05342578549526981\n",
+ "key 1 value[2] 0.1466022529006109\n",
+ "key 2 value[2] 0.2581001449566536\n",
+ "key 3 value[2] 0.4070438657500979\n",
+ "key 4 value[2] 0.6203607283161051\n",
+ "key 5 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07176509439127023\n",
+ "Current Error err_H 0.0888887469703822\n",
+ "myFeeder.t 5\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00923444 0.01113374 ... 0.01842175 0. 0. ]\n",
+ " [0. 0.00834078 0.01167685 ... 0.01778833 0. 0. ]\n",
+ " [0. 0.00834078 0.01167685 ... 0.01778833 0. 0. ]\n",
+ " ...\n",
+ " [0.50788235 0.35275544 0.32050238 ... 0.02908419 0.13381529 0.44933687]\n",
+ " [0.54630955 0.36776884 0.31690428 ... 0.02855635 0.12861016 0.46330681]\n",
+ " [0.55878432 0.37155198 0.31059063 ... 0.02813407 0.12313995 0.46595933]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.65981531e-04 2.28784793e-02 ... 1.87384534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.00950441e-02 ... 1.84745315e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.96877122e-02 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.90166942e-01 3.51921358e-01 3.28716904e-01 ... 2.77117973e-02\n",
+ " 1.89312609e-01 5.01503095e-01]\n",
+ " [5.84638904e-01 3.52874590e-01 3.38628649e-01 ... 2.77117973e-02\n",
+ " 1.88204109e-01 5.01326260e-01]\n",
+ " [6.27848101e-01 3.53500149e-01 3.31093007e-01 ... 2.97703880e-02\n",
+ " 1.82203747e-01 4.94960212e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 6.25558534e-04 2.65784114e-02 ... 1.89231987e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 6.25558534e-04 2.65784114e-02 ... 1.89231987e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 6.25558534e-04 2.65784114e-02 ... 1.89231987e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.94037791e-01 3.68364611e-01 3.31873727e-01 ... 3.04829770e-02\n",
+ " 1.90475330e-01 5.24137931e-01]\n",
+ " [5.17739864e-01 3.58221626e-01 3.08757637e-01 ... 2.81868567e-02\n",
+ " 1.44279776e-01 5.10344828e-01]\n",
+ " [6.94496423e-01 3.76184093e-01 3.45824847e-01 ... 3.23832146e-02\n",
+ " 1.92264594e-01 5.27320955e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.36193029e-04 2.57637475e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 2.55600815e-02 ... 1.86856690e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.52545825e-02 ... 1.86856690e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [7.10530178e-01 3.57551385e-01 3.02851324e-01 ... 3.23040380e-02\n",
+ " 1.79215615e-01 5.04509284e-01]\n",
+ " [7.45661347e-01 3.39365505e-01 2.87372709e-01 ... 3.13539192e-02\n",
+ " 1.92065787e-01 4.77984085e-01]\n",
+ " [7.51770317e-01 3.61394102e-01 3.41140530e-01 ... 3.53919240e-02\n",
+ " 1.91378999e-01 5.48010610e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.76616026e-04 1.77189409e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.83978276e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.98913781e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.94478077e-01 3.60023831e-01 3.38085540e-01 ... 3.56822381e-02\n",
+ " 1.97602265e-01 4.90893015e-01]\n",
+ " [6.92961536e-01 3.57223712e-01 3.28309572e-01 ... 3.54183162e-02\n",
+ " 1.96385324e-01 4.80636605e-01]\n",
+ " [6.57555189e-01 3.58951445e-01 3.37270876e-01 ... 3.40987068e-02\n",
+ " 1.96325080e-01 4.29354553e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.06404528e-04 2.54582485e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.54582485e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.54582485e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [7.31670030e-01 3.35358951e-01 3.32315003e-01 ... 3.54183162e-02\n",
+ " 1.96614254e-01 4.41025641e-01]\n",
+ " [7.71601541e-01 3.56419422e-01 3.44399185e-01 ... 3.13539192e-02\n",
+ " 1.94505693e-01 4.35720601e-01]\n",
+ " [7.48535437e-01 3.42389038e-01 3.40325866e-01 ... 3.15122724e-02\n",
+ " 1.79312007e-01 4.29177719e-01]]\n",
+ "list_RMSE_SousModele [0.08695765854633636, 0.07362281360063773, 0.09431166690925853, 0.08179394197647431, 0.08214175379134846, 0.05384648929203804] mean of list_RMSE_SousModele 0.07877905401934891\n",
+ " RMSE resultat vote 0.057653976383557515\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "key 0 value[6] 0.18670485439918758\n",
+ "key 1 value[6] 0.21615527050978905\n",
+ "key 2 value[6] 0.25465259711919624\n",
+ "key 3 value[6] 0.3081780004228992\n",
+ "key 4 value[6] 0.39107838594488675\n",
+ "key 5 value[6] 0.5474158342212556\n",
+ "key 6 value[6] 1.0\n",
+ "key 0 value[2] 0.04957382864696816\n",
+ "key 1 value[2] 0.12396886473608414\n",
+ "key 2 value[2] 0.2102976069731129\n",
+ "key 3 value[2] 0.3192541078092026\n",
+ "key 4 value[2] 0.4604164119139006\n",
+ "key 5 value[2] 0.6627833895596708\n",
+ "key 6 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06941324138998477\n",
+ "Current Error err_H 0.057653976383557515\n",
+ "myFeeder.t 6\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.54961169 0.37387548 0.30794297 ... 0.02813407 0.12309175 0.46825818]\n",
+ " [0.44204733 0.35918975 0.32084182 ... 0.02934811 0.12633291 0.45747126]\n",
+ " [0.46435516 0.35889187 0.32118126 ... 0.02913698 0.14024941 0.45499558]\n",
+ " ...\n",
+ " [0. 0.1010426 0.06904277 ... 0.03304302 0. 0.05923961]\n",
+ " [0. 0.05245755 0.04534963 ... 0.02955925 0. 0.02316534]\n",
+ " [0. 0.07476914 0.05906314 ... 0.02612827 0. 0.04615385]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.58345258 0.35251713 0.34175153 ... 0.02713117 0.18803542 0.49496021]\n",
+ " [0.60374243 0.35117665 0.33102512 ... 0.02744788 0.18231219 0.49867374]\n",
+ " [0.58211949 0.35582365 0.34209097 ... 0.02845078 0.18132418 0.50044209]\n",
+ " ...\n",
+ " [0. 0.08087578 0.05485404 ... 0.02164159 0. 0.03607427]\n",
+ " [0. 0.0515639 0.05302105 ... 0.0202692 0. 0. ]\n",
+ " [0. 0.11081323 0.0736592 ... 0.02370018 0. 0.04049514]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.09099248e-01 3.60277033e-01 3.05804481e-01 ... 3.19081552e-02\n",
+ " 1.71371769e-01 5.18302387e-01]\n",
+ " [6.48064575e-01 3.63672922e-01 3.10488798e-01 ... 3.25415677e-02\n",
+ " 1.80607265e-01 5.16976127e-01]\n",
+ " [5.41148413e-01 3.62019660e-01 3.06517312e-01 ... 2.65241489e-02\n",
+ " 1.78908368e-01 5.12201592e-01]\n",
+ " ...\n",
+ " [3.13703908e-03 1.41867739e-01 7.90224033e-02 ... 2.17735550e-02\n",
+ " 5.42201337e-04 1.73209549e-01]\n",
+ " [0.00000000e+00 1.73056300e-01 5.85539715e-02 ... 3.08788599e-02\n",
+ " 0.00000000e+00 1.18302387e-01]\n",
+ " [8.42047331e-03 8.47184987e-02 9.78615071e-02 ... 2.45447348e-02\n",
+ " 4.01228990e-03 1.23872679e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.72228949 0.34468275 0.30193483 ... 0.03230404 0.19163203 0.49920424]\n",
+ " [0.65690699 0.35571939 0.31904277 ... 0.03269992 0.19423459 0.50185676]\n",
+ " [0.59849569 0.36000894 0.34348269 ... 0.03499604 0.19569854 0.51909814]\n",
+ " ...\n",
+ " [0. 0.08208222 0.02189409 ... 0.02304038 0. 0.14297082]\n",
+ " [0. 0.03087578 0.048778 ... 0.02050673 0. 0.01193634]\n",
+ " [0. 0.09789991 0.0904277 ... 0.02660333 0. 0.01220159]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.65755519 0.35895144 0.33727088 ... 0.03409871 0.19632508 0.42935455]\n",
+ " [0.6175503 0.36032172 0.32980312 ... 0.03430984 0.19425267 0.44756852]\n",
+ " [0.59320002 0.36026214 0.33849287 ... 0.03077329 0.19696367 0.49531388]\n",
+ " ...\n",
+ " [0. 0.01912422 0.02437203 ... 0.02185273 0. 0.04562334]\n",
+ " [0. 0.08084599 0.04928717 ... 0.02950647 0. 0.1734748 ]\n",
+ " [0. 0.0629431 0.07033265 ... 0.02596991 0. 0.0489832 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.65360484 0.35886208 0.36829599 ... 0.03072051 0.19351768 0.43607427]\n",
+ " [0.61232801 0.35537682 0.36429056 ... 0.02929533 0.1950238 0.43908046]\n",
+ " [0.5939216 0.35019363 0.34446707 ... 0.03161784 0.17816736 0.45004421]\n",
+ " ...\n",
+ " [0.00496545 0.01843908 0.04365241 ... 0.02528372 0.00083138 0. ]\n",
+ " [0.02721213 0.03696753 0.05424304 ... 0.02639219 0.00261462 0.00495137]\n",
+ " [0. 0.02213286 0.03937542 ... 0.02512536 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.63798691 0.37515639 0.36055669 ... 0.03109 0.19192722 0.43147657]\n",
+ " [0.57618786 0.37336908 0.35790903 ... 0.03235682 0.19178264 0.43271441]\n",
+ " [0.61629059 0.37515639 0.36632722 ... 0.03109 0.18424001 0.42581786]\n",
+ " ...\n",
+ " [0. 0.07736074 0.067685 ... 0.02401689 0. 0.00795756]\n",
+ " [0. 0.07542449 0.06463001 ... 0.02443917 0. 0. ]\n",
+ " [0. 0.03351206 0.03856076 ... 0.02153603 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.09087084346660503, 0.07944626657520108, 0.09957602963533363, 0.08920171367660132, 0.08707908957509854, 0.06886710438381921, 0.0592504075888797] mean of list_RMSE_SousModele 0.08204163641450549\n",
+ " RMSE resultat vote 0.05978007937397294\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 0 value[6] 0.16334850821192484\n",
+ "key 1 value[6] 0.18670485439918758\n",
+ "key 2 value[6] 0.21615527050978905\n",
+ "key 3 value[6] 0.25465259711919624\n",
+ "key 4 value[6] 0.3081780004228992\n",
+ "key 5 value[6] 0.39107838594488675\n",
+ "key 6 value[6] 0.5474158342212556\n",
+ "key 7 value[6] 1.0\n",
+ "key 0 value[2] 0.04520106778517789\n",
+ "key 1 value[2] 0.10452605990206913\n",
+ "key 2 value[2] 0.17202675539635087\n",
+ "key 3 value[2] 0.25424647528976846\n",
+ "key 4 value[2] 0.3558935887665646\n",
+ "key 5 value[2] 0.4882277262566839\n",
+ "key 6 value[2] 0.6729614108620893\n",
+ "key 7 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06803707538769736\n",
+ "Current Error err_H 0.05978007937397294\n",
+ "myFeeder.t 7\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.09609771 0.06293279 ... 0.03225125 0. 0.04067197]\n",
+ " [0. 0.08150134 0.0539036 ... 0.02918976 0. 0.03023873]\n",
+ " [0. 0.10539172 0.06429056 ... 0.03003431 0. 0.04314766]\n",
+ " ...\n",
+ " [0. 0.01313673 0.01025119 ... 0.01815783 0. 0. ]\n",
+ " [0. 0.01313673 0.01025119 ... 0.01831618 0. 0. ]\n",
+ " [0. 0.01313673 0.01025119 ... 0.01831618 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.09862973 0.06171079 ... 0.02501979 0. 0.06525199]\n",
+ " [0. 0.03997617 0.04344874 ... 0.02660333 0. 0. ]\n",
+ " [0. 0.00643432 0.02973523 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.13025022 0.07158859 ... 0.02438638 0. 0.22732095]\n",
+ " [0. 0.0528597 0.02606925 ... 0.0229612 0. 0.14350133]\n",
+ " [0. 0.07689902 0.02973523 ... 0.02454473 0. 0.24005305]\n",
+ " ...\n",
+ " [0. 0.0075067 0.02199593 ... 0.01955661 0. 0.04854111]\n",
+ " [0. 0.0075067 0.02199593 ... 0.01955661 0. 0.04854111]\n",
+ " [0. 0.00062556 0.02138493 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.12913315 0.04246436 ... 0.02850356 0. 0.24429708]\n",
+ " [0. 0.09874888 0.03533605 ... 0.02383215 0. 0.24323607]\n",
+ " [0. 0.01470063 0.00804481 ... 0.02304038 0. 0.12599469]\n",
+ " ...\n",
+ " [0. 0.00866845 0.01232179 ... 0.02177356 0. 0.0729443 ]\n",
+ " [0. 0.00866845 0.01232179 ... 0.02185273 0. 0.0729443 ]\n",
+ " [0. 0.00866845 0.01232179 ... 0.02185273 0. 0.0729443 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.02114984 0.03441955 ... 0.02443917 0. 0.02528736]\n",
+ " [0. 0.02147751 0.02600136 ... 0.02206387 0. 0.03076923]\n",
+ " [0. 0.00044683 0.02410048 ... 0.02216944 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02267481 ... 0.02253893 0. 0. ]\n",
+ " [0. 0.00044683 0.02267481 ... 0.02253893 0. 0. ]\n",
+ " [0. 0.00044683 0.02267481 ... 0.02253893 0. 0. ]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.02472446 0.04534963 ... 0.02385854 0. 0. ]\n",
+ " [0.00528343 0.02695859 0.04494229 ... 0.02422803 0.00087957 0. ]\n",
+ " [0. 0.02886506 0.05125594 ... 0.02781737 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03109301 ... 0.02755344 0. 0. ]\n",
+ " [0. 0.00044683 0.03109301 ... 0.02755344 0. 0. ]\n",
+ " [0. 0.00041704 0.02953157 ... 0.02686725 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00448847 0.07712243 0.08438561 ... 0.02607548 0.00074703 0.00477454]\n",
+ " [0.00913594 0.08987191 0.08465716 ... 0.02692003 0.00150611 0.01043324]\n",
+ " [0. 0.10157879 0.06985743 ... 0.02380575 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00187668 0.02668024 ... 0.01989971 0. 0. ]\n",
+ " [0. 0.00187668 0.02668024 ... 0.01989971 0. 0. ]\n",
+ " [0. 0.00140006 0.02681602 ... 0.01995249 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.05607685 0.04714868 ... 0.02359462 0. 0. ]\n",
+ " [0.02091359 0.08842717 0.07433809 ... 0.02438638 0. 0.03342175]\n",
+ " [0. 0.05089366 0.0385947 ... 0.0216152 0. 0.01671088]\n",
+ " ...\n",
+ " [0. 0.02841823 0.04511202 ... 0.02050673 0. 0.01061008]\n",
+ " [0. 0.02841823 0.04541752 ... 0.02050673 0. 0.01061008]\n",
+ " [0. 0.02841823 0.04541752 ... 0.02050673 0. 0.01061008]]\n",
+ "list_RMSE_SousModele [0.09853296839447885, 0.07949888136018943, 0.09217167403568766, 0.08406303793258235, 0.08137288425406843, 0.0790937891048099, 0.07122947491886253, 0.0715724879458633] mean of list_RMSE_SousModele 0.08219189974331781\n",
+ " RMSE resultat vote 0.06264369230976617\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 0 value[6] 0.1444116730248319\n",
+ "key 1 value[6] 0.16334850821192484\n",
+ "key 2 value[6] 0.18670485439918758\n",
+ "key 3 value[6] 0.21615527050978905\n",
+ "key 4 value[6] 0.25465259711919624\n",
+ "key 5 value[6] 0.3081780004228992\n",
+ "key 6 value[6] 0.39107838594488675\n",
+ "key 7 value[6] 0.5474158342212556\n",
+ "key 8 value[6] 1.0\n",
+ "key 0 value[2] 0.04432274678060647\n",
+ "key 1 value[2] 0.09495675713928708\n",
+ "key 2 value[2] 0.15111015872353975\n",
+ "key 3 value[2] 0.21683079318274517\n",
+ "key 4 value[2] 0.2946060593289635\n",
+ "key 5 value[2] 0.38947960288888234\n",
+ "key 6 value[2] 0.51227092589013\n",
+ "key 7 value[2] 0.6855644067882606\n",
+ "key 8 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06736290250295596\n",
+ "Current Error err_H 0.06264369230976617\n",
+ "myFeeder.t 8\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01313673 0.01025119 ... 0.01831618 0. 0. ]\n",
+ " [0. 0.01313673 0.01025119 ... 0.01815783 0. 0. ]\n",
+ " [0. 0.01313673 0.01025119 ... 0.01815783 0. 0. ]\n",
+ " ...\n",
+ " [0.54358222 0.35355973 0.32138493 ... 0.03035102 0.11592265 0.43925729]\n",
+ " [0.53736929 0.35281501 0.31860149 ... 0.03029823 0.12793542 0.43625111]\n",
+ " [0.54061029 0.35758117 0.3095723 ... 0.02860913 0.13321284 0.41609195]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.0041406 0.02817379 ... 0.01847453 0. 0. ]\n",
+ " ...\n",
+ " [0.58944536 0.3536193 0.33021045 ... 0.02792293 0.18867402 0.50044209]\n",
+ " [0.60657983 0.35171284 0.3353021 ... 0.02787015 0.18210736 0.49708223]\n",
+ " [0.60560142 0.32892464 0.32443992 ... 0.02829243 0.18937285 0.46878868]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 6.25558534e-04 2.13849287e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 6.25558534e-04 2.13849287e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 6.25558534e-04 2.13849287e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [7.08824069e-01 3.66800715e-01 3.18635438e-01 ... 3.14330958e-02\n",
+ " 1.80354238e-01 5.16445623e-01]\n",
+ " [6.34672537e-01 3.61840929e-01 2.95112016e-01 ... 3.15122724e-02\n",
+ " 1.68190855e-01 5.21750663e-01]\n",
+ " [7.45477894e-01 3.58355675e-01 3.25661914e-01 ... 3.09580364e-02\n",
+ " 1.95174408e-01 4.92042440e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00701519 0.00773931 ... 0.02050673 0. 0.05835544]\n",
+ " [0. 0.00701519 0.00773931 ... 0.02050673 0. 0.05835544]\n",
+ " [0. 0.00701519 0.00773931 ... 0.02050673 0. 0.05835544]\n",
+ " ...\n",
+ " [0.66941846 0.32394996 0.22138493 ... 0.02699921 0.17442617 0.41618037]\n",
+ " [0.71506146 0.30424486 0.19898167 ... 0.02652415 0.13602024 0.38143236]\n",
+ " [0.76222711 0.33838248 0.31619145 ... 0.03056215 0.19725285 0.48912467]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.26748133e-02 ... 2.25389285e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.26748133e-02 ... 2.25389285e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.26748133e-02 ... 2.25389285e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [7.04665811e-01 3.05064045e-01 3.03394433e-01 ... 3.17234099e-02\n",
+ " 1.82818242e-01 4.64544651e-01]\n",
+ " [6.98770868e-01 3.46708371e-01 3.36320434e-01 ... 3.28318818e-02\n",
+ " 1.95409362e-01 4.68258179e-01]\n",
+ " [7.44046964e-01 3.58921656e-01 3.36048880e-01 ... 3.13011349e-02\n",
+ " 1.96060004e-01 4.86295314e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.17039023e-04 2.95315682e-02 ... 2.68672473e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.17039023e-04 2.95315682e-02 ... 2.68672473e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.17039023e-04 2.95315682e-02 ... 2.68672473e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.67596160e-01 3.52427763e-01 3.49490835e-01 ... 3.28846661e-02\n",
+ " 1.61021748e-01 4.64898320e-01]\n",
+ " [6.69235003e-01 3.37354781e-01 3.32179226e-01 ... 3.24096068e-02\n",
+ " 1.55431050e-01 4.56056587e-01]\n",
+ " [6.83654375e-01 3.50372356e-01 3.49898167e-01 ... 3.91660069e-02\n",
+ " 1.76745587e-01 4.59239611e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00140006 0.02681602 ... 0.01995249 0. 0. ]\n",
+ " [0. 0.00187668 0.02668024 ... 0.01989971 0. 0. ]\n",
+ " [0. 0.00187668 0.02668024 ... 0.01989971 0. 0. ]\n",
+ " ...\n",
+ " [0.70634134 0.35498957 0.34290563 ... 0.03124835 0.17358877 0.42670203]\n",
+ " [0.6956644 0.35877271 0.35573659 ... 0.03177619 0.17299837 0.42723254]\n",
+ " [0.69187305 0.35656836 0.35539715 ... 0.03193455 0.17527562 0.41679929]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.02609473 0.04521385 ... 0.02050673 0. 0.02732095]\n",
+ " [0. 0.02841823 0.04511202 ... 0.02050673 0. 0.01061008]\n",
+ " [0. 0.02846291 0.04551935 ... 0.02082344 0. 0.01061008]\n",
+ " ...\n",
+ " [0.69258852 0.33588025 0.3308554 ... 0.03159145 0.18031809 0.45809019]\n",
+ " [0.72423409 0.34271671 0.3101833 ... 0.03428345 0.16696187 0.43156499]\n",
+ " [0.62280316 0.31242181 0.26069246 ... 0.03317498 0.15881077 0.3464191 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 3.97827563e-02 ... 2.25917129e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.96469790e-02 ... 2.26972816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.95112016e-02 ... 2.25917129e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.88081698e-01 3.53053321e-01 3.50848608e-01 ... 3.33069411e-02\n",
+ " 9.28851136e-02 4.27055703e-01]\n",
+ " [7.01400355e-01 3.41971999e-01 3.37067210e-01 ... 3.48904724e-02\n",
+ " 9.29333092e-02 4.21927498e-01]\n",
+ " [7.31963554e-01 3.48853143e-01 3.49626612e-01 ... 3.30430193e-02\n",
+ " 1.04982228e-01 4.37135279e-01]]\n",
+ "list_RMSE_SousModele [0.09889219440801612, 0.08219104802288275, 0.09252212850047901, 0.08493716546434225, 0.08022931611062804, 0.09150294902243566, 0.07943284724953509, 0.07664852208645836, 0.06071984321764075] mean of list_RMSE_SousModele 0.08300844600915758\n",
+ " RMSE resultat vote 0.06201755783195607\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "key 0 value[6] 0.1288294285628787\n",
+ "key 1 value[6] 0.1444116730248319\n",
+ "key 2 value[6] 0.16334850821192484\n",
+ "key 3 value[6] 0.18670485439918758\n",
+ "key 4 value[6] 0.21615527050978905\n",
+ "key 5 value[6] 0.25465259711919624\n",
+ "key 6 value[6] 0.3081780004228992\n",
+ "key 7 value[6] 0.39107838594488675\n",
+ "key 8 value[6] 0.5474158342212556\n",
+ "key 9 value[6] 1.0\n",
+ "key 0 value[2] 0.04216770276082402\n",
+ "key 1 value[2] 0.08527322375274034\n",
+ "key 2 value[2] 0.1325873983544471\n",
+ "key 3 value[2] 0.18697225427688607\n",
+ "key 4 value[2] 0.25000506355358837\n",
+ "key 5 value[2] 0.32401077023058444\n",
+ "key 6 value[2] 0.41605894972925256\n",
+ "key 7 value[2] 0.5353968097327484\n",
+ "key 8 value[2] 0.70906615420327\n",
+ "key 9 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.0667689753172893\n",
+ "Current Error err_H 0.06201755783195607\n",
+ "myFeeder.t 9\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.51556289 0.34867441 0.30380177 ... 0.02787015 0.12788722 0.4183908 ]\n",
+ " [0.46200697 0.35522788 0.31860149 ... 0.02839799 0.12794747 0.44969054]\n",
+ " [0.46200697 0.35522788 0.31860149 ... 0.02839799 0.12794747 0.44969054]\n",
+ " ...\n",
+ " [0. 0.00947274 0.01221996 ... 0.01762998 0. 0. ]\n",
+ " [0. 0.01128984 0.01120163 ... 0.01752441 0. 0. ]\n",
+ " [0. 0.01039619 0.01188052 ... 0.01773555 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.33486210e-01 3.23056300e-01 3.15071283e-01 ... 2.87674848e-02\n",
+ " 1.89927104e-01 4.57471264e-01]\n",
+ " [7.36843393e-01 3.35597259e-01 3.43448744e-01 ... 3.44154130e-02\n",
+ " 1.86673896e-01 4.63837312e-01]\n",
+ " [7.36843393e-01 3.35597259e-01 3.43448744e-01 ... 3.44154130e-02\n",
+ " 1.86673896e-01 4.63837312e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.36193029e-04 1.46639511e-02 ... 1.77355503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 1.45960625e-02 ... 1.76827659e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 1.46639511e-02 ... 1.77355503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.63804806e-01 3.39008043e-01 3.14460285e-01 ... 2.87410926e-02\n",
+ " 1.96276884e-01 4.55437666e-01]\n",
+ " [7.61108054e-01 3.38069705e-01 3.08757637e-01 ... 2.89786223e-02\n",
+ " 1.96041930e-01 4.54641910e-01]\n",
+ " [7.61841864e-01 3.34450402e-01 3.00814664e-01 ... 2.98495645e-02\n",
+ " 1.95861196e-01 4.50663130e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.47454175e-02 ... 1.82897862e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.47454175e-02 ... 1.82897862e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.80875782e-04 2.34215886e-02 ... 1.84481394e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.82975601e-01 3.36282395e-01 3.23930754e-01 ... 3.04829770e-02\n",
+ " 1.98608350e-01 4.89920424e-01]\n",
+ " [7.79361585e-01 3.25290438e-01 3.09572301e-01 ... 3.00079177e-02\n",
+ " 1.98156515e-01 4.55437666e-01]\n",
+ " [7.79655109e-01 3.03574620e-01 2.87780041e-01 ... 2.89786223e-02\n",
+ " 2.01500090e-01 4.60742706e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.91510277e-04 9.16496945e-03 ... 1.81314331e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 9.16496945e-03 ... 1.81314331e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 2.10008937e-03 7.23014257e-03 ... 1.81314331e-02\n",
+ " 0.00000000e+00 6.36604775e-03]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.47618174e-01 3.50193625e-01 3.38560760e-01 ... 3.13539192e-02\n",
+ " 1.94951503e-01 4.86472149e-01]\n",
+ " [7.56215985e-01 3.52100089e-01 3.26816022e-01 ... 3.43626287e-02\n",
+ " 1.93806856e-01 4.44916004e-01]\n",
+ " [7.53769950e-01 3.55704498e-01 3.26069246e-01 ... 3.51016099e-02\n",
+ " 1.93481535e-01 4.40318302e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.08418194e-02 ... 2.10609660e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.02308215e-02 ... 2.10081816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.08418194e-02 ... 2.12721035e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.83140708e-01 3.49865952e-01 3.51323829e-01 ... 3.21984693e-02\n",
+ " 1.84830411e-01 4.49336870e-01]\n",
+ " [7.26863572e-01 3.40035746e-01 3.76985743e-01 ... 2.96120348e-02\n",
+ " 1.94397253e-01 4.12908930e-01]\n",
+ " [7.29003853e-01 3.36103664e-01 3.72437203e-01 ... 2.96120348e-02\n",
+ " 1.94361106e-01 4.12732095e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.06404528e-04 2.53903598e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.53224711e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.53903598e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.64634012 0.34366994 0.34317719 ... 0.03272631 0.17814326 0.40548187]\n",
+ " [0.74323977 0.35382782 0.36245757 ... 0.02760623 0.18922827 0.43501326]\n",
+ " [0.73428729 0.35394698 0.35682281 ... 0.02797572 0.18981866 0.43483643]\n",
+ " ...\n",
+ " [0. 0.00354483 0.02471147 ... 0.01974136 0. 0. ]\n",
+ " [0. 0.00354483 0.02471147 ... 0.01974136 0. 0. ]\n",
+ " [0. 0.00354483 0.02471147 ... 0.01974136 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.65125665e-01 3.50983021e-01 3.41751527e-01 ... 3.46001584e-02\n",
+ " 1.50515091e-01 4.00265252e-01]\n",
+ " [6.52228949e-01 3.53887399e-01 3.27189409e-01 ... 3.53919240e-02\n",
+ " 1.60021688e-01 4.07161804e-01]\n",
+ " [7.50486149e-01 3.58400357e-01 3.35234216e-01 ... 3.48376880e-02\n",
+ " 1.91288632e-01 4.14323607e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.56619145e-02 ... 1.87648456e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 1.11260054e-02 2.54582485e-02 ... 1.87648456e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.56619145e-02 ... 1.91607284e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.39277197e-01 3.46738159e-01 3.40325866e-01 ... 3.28846661e-02\n",
+ " 9.93794807e-02 4.33244916e-01]\n",
+ " [7.15232679e-01 3.53321418e-01 3.59809912e-01 ... 3.30430193e-02\n",
+ " 9.20657871e-02 4.37842617e-01]\n",
+ " [7.11355715e-01 3.59457849e-01 3.62525458e-01 ... 3.42042755e-02\n",
+ " 9.28730646e-02 4.37842617e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 2.70875764e-02 ... 1.88440222e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.66802444e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.70875764e-02 ... 1.89495909e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.75521311e-01 3.40661305e-01 3.26476578e-01 ... 3.34125099e-02\n",
+ " 1.79769866e-01 4.04951370e-01]\n",
+ " [7.32489451e-01 3.64045279e-01 3.38900204e-01 ... 3.52071787e-02\n",
+ " 1.86782336e-01 4.25464191e-01]\n",
+ " [7.35400232e-01 3.66904975e-01 3.40054311e-01 ... 3.56822381e-02\n",
+ " 1.81107296e-01 4.25994695e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.07535642e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.03462322e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.00067889e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.10490800829198102, 0.07426611313903203, 0.08910612402558585, 0.078429512031004, 0.07495233259379147, 0.08346573841816639, 0.08414067215562925, 0.08711038413586951, 0.06425136765363525, 0.05810034485844437] mean of list_RMSE_SousModele 0.07987305973031392\n",
+ " RMSE resultat vote 0.05860493048519744\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "key 0 value[6] 0.11586412658816056\n",
+ "key 1 value[6] 0.1288294285628787\n",
+ "key 2 value[6] 0.1444116730248319\n",
+ "key 3 value[6] 0.16334850821192484\n",
+ "key 4 value[6] 0.18670485439918758\n",
+ "key 5 value[6] 0.21615527050978905\n",
+ "key 6 value[6] 0.25465259711919624\n",
+ "key 7 value[6] 0.3081780004228992\n",
+ "key 8 value[6] 0.39107838594488675\n",
+ "key 9 value[6] 0.5474158342212556\n",
+ "key 10 value[6] 1.0\n",
+ "key 0 value[2] 0.038833008019364304\n",
+ "key 1 value[2] 0.07552370627375947\n",
+ "key 2 value[2] 0.1150061799376607\n",
+ "key 3 value[2] 0.15979436577521436\n",
+ "key 4 value[2] 0.21057607187113384\n",
+ "key 5 value[2] 0.26899783977353153\n",
+ "key 6 value[2] 0.33918619431585784\n",
+ "key 7 value[2] 0.4261833539350207\n",
+ "key 8 value[2] 0.5438244017110664\n",
+ "key 9 value[2] 0.7051907063863734\n",
+ "key 10 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06595257083408011\n",
+ "Current Error err_H 0.05860493048519744\n",
+ "myFeeder.t 10\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.01459606 ... 0.0175772 0. 0. ]\n",
+ " [0. 0.0005064 0.01466395 ... 0.0175772 0. 0. ]\n",
+ " [0. 0.00047662 0.01568228 ... 0.0175772 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00226393 0.01955193 ... 0.01810504 0. 0. ]\n",
+ " [0. 0.00047662 0.01907671 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01446029 ... 0.01794669 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.02260692 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00049151 0.02087576 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00058088 0.0206721 ... 0.01718131 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00049151 0.02474542 ... 0.0192399 0. 0. ]\n",
+ " [0. 0.00044683 0.02464358 ... 0.0202692 0. 0. ]\n",
+ " [0. 0.02618409 0.03421589 ... 0.02034838 0. 0.03448276]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00210009 0.00947047 ... 0.0178939 0. 0.00636605]\n",
+ " [0. 0.00049151 0.00539715 ... 0.01781473 0. 0. ]\n",
+ " [0. 0.0023235 0.00916497 ... 0.0178939 0. 0.01405836]\n",
+ " ...\n",
+ " [0. 0.00062556 0.00936864 ... 0.01828979 0. 0. ]\n",
+ " [0. 0.00652368 0.00804481 ... 0.01844814 0. 0.0535809 ]\n",
+ " [0. 0.00214477 0.00712831 ... 0.01821061 0. 0.01193634]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02023082 ... 0.02100818 0. 0. ]\n",
+ " [0. 0.00044683 0.02267481 ... 0.02206387 0. 0. ]\n",
+ " [0. 0.00041704 0.02233537 ... 0.0219583 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.01111111 0.01588595 ... 0.0219583 0. 0.05729443]\n",
+ " [0. 0.00253202 0.0185336 ... 0.01989971 0. 0. ]\n",
+ " [0. 0.00044683 0.02002716 ... 0.0192663 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02600136 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00047662 0.02484725 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00047662 0.02661236 ... 0.02148324 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00250223 0.02579769 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.00253202 0.02579769 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00047662 0.02443992 ... 0.01836896 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00953232 0.03143245 ... 0.02016363 0. 0. ]\n",
+ " [0. 0.01581769 0.03842498 ... 0.02090261 0. 0. ]\n",
+ " [0. 0.00351504 0.02532247 ... 0.02000528 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00119154 0.02532247 ... 0.02000528 0. 0. ]\n",
+ " [0. 0.00092344 0.02511881 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.00107239 0.02559403 ... 0.02005806 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02586558 ... 0.0192399 0. 0. ]\n",
+ " [0. 0.00044683 0.02535642 ... 0.01876485 0. 0. ]\n",
+ " [0. 0.00044683 0.02525458 ... 0.01916073 0. 0. ]\n",
+ " ...\n",
+ " [0.00961292 0.02596068 0.02617108 ... 0.01900238 0.00131936 0. ]\n",
+ " [0. 0.02198391 0.02708758 ... 0.01876485 0. 0. ]\n",
+ " [0. 0.00044683 0.02657841 ... 0.01868567 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02776646 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.00047662 0.02681602 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02579769 ... 0.01894959 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.02365207 0.03217923 ... 0.01826339 0. 0. ]\n",
+ " [0. 0.03190349 0.03564155 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.017605 0.03048201 ... 0.01842175 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03279022 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00053619 0.03272234 ... 0.01963579 0. 0. ]\n",
+ " [0. 0.00065535 0.0365241 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.04980637 0.06551256 ... 0.02586434 0. 0.00229885]\n",
+ " [0. 0.06767948 0.08811948 ... 0.01984693 0. 0.00725022]\n",
+ " [0. 0.04813822 0.05790903 ... 0.01905516 0. 0.00530504]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02953157 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02749491 ... 0.02612827 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.0305499 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.02647658 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.0305499 ... 0.01821061 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.07435809303578006, 0.0905336272288401, 0.07967301012640585, 0.07456459599228987, 0.08356203290129799, 0.0848929769409826, 0.08647007232857062, 0.06288729506280866, 0.05676351060294694, 0.07892983962829869] mean of list_RMSE_SousModele 0.07726350538482214\n",
+ " RMSE resultat vote 0.05515201309621722\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 1 value[6] 0.11586412658816056\n",
+ "key 2 value[6] 0.1288294285628787\n",
+ "key 3 value[6] 0.1444116730248319\n",
+ "key 4 value[6] 0.16334850821192484\n",
+ "key 5 value[6] 0.18670485439918758\n",
+ "key 6 value[6] 0.21615527050978905\n",
+ "key 7 value[6] 0.25465259711919624\n",
+ "key 8 value[6] 0.3081780004228992\n",
+ "key 9 value[6] 0.39107838594488675\n",
+ "key 10 value[6] 0.5474158342212556\n",
+ "key 11 value[6] 1.0\n",
+ "key 1 value[2] 0.04034939314734483\n",
+ "key 2 value[2] 0.07723311329201783\n",
+ "key 3 value[2] 0.11856032261041596\n",
+ "key 4 value[2] 0.1647144331004789\n",
+ "key 5 value[2] 0.21673883851002435\n",
+ "key 6 value[2] 0.2774231060956513\n",
+ "key 7 value[2] 0.34969199080507174\n",
+ "key 8 value[2] 0.4401570420449772\n",
+ "key 9 value[2] 0.5522301530268416\n",
+ "key 10 value[2] 0.7103877600445895\n",
+ "key 11 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06497070194881985\n",
+ "Current Error err_H 0.05515201309621722\n",
+ "myFeeder.t 11\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00137027 0.01527495 ... 0.01784112 0. 0. ]\n",
+ " [0. 0.00044683 0.01500339 ... 0.01799947 0. 0. ]\n",
+ " [0. 0.0049151 0.02179226 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02980312 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02973523 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02973523 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00214477 0.00936864 ... 0.01828979 0. 0. ]\n",
+ " [0. 0.00049151 0.0095723 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00044683 0.0114053 ... 0.01852732 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00053619 0.01456212 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00053619 0.01446029 ... 0.01844814 0. 0. ]\n",
+ " [0. 0.00053619 0.01456212 ... 0.01852732 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00056598 0.02084182 ... 0.01889681 0. 0. ]\n",
+ " [0. 0.00041704 0.02002716 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.00044683 0.02050238 ... 0.01931908 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0005064 0.0221317 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.0221317 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.0221317 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02511881 ... 0.01836896 0. 0. ]\n",
+ " [0. 0.00047662 0.02511881 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00253202 0.02627291 ... 0.01847453 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02566191 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01274948 0.0371351 ... 0.02037477 0. 0. ]\n",
+ " [0. 0.00381293 0.02498303 ... 0.01863288 0. 0. ]\n",
+ " [0. 0.0028597 0.02532247 ... 0.01879124 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.02593347 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02579769 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02586558 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.02596741 ... 0.01805226 0. 0. ]\n",
+ " [0. 0.00049151 0.02739308 ... 0.01844814 0. 0. ]\n",
+ " [0. 0.01117069 0.02698574 ... 0.01781473 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03401222 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03431772 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03411405 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01757522 0.03095723 ... 0.01799947 0. 0. ]\n",
+ " [0. 0.01611558 0.02837746 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.01974978 0.02817379 ... 0.01805226 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.0269518 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.02701969 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02674813 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.03410783 0.04487441 ... 0.01900238 0. 0.00229885]\n",
+ " [0.00540574 0.04986595 0.04738629 ... 0.01847453 0. 0.00548187]\n",
+ " [0. 0.02597557 0.04209097 ... 0.01852732 0. 0.00229885]\n",
+ " ...\n",
+ " [0. 0.00047662 0.03380855 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00053619 0.03251867 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.03380855 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.04959786 0.06822811 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.02545825 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.03156823 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00089366 0.02953157 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00089366 0.02953157 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00089366 0.02953157 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02640869 ... 0.02338348 0. 0. ]\n",
+ " [0. 0.00047662 0.02613714 ... 0.02385854 0. 0. ]\n",
+ " [0. 0.00047662 0.02545825 ... 0.02359462 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0005064 0.02776646 ... 0.02153603 0. 0. ]\n",
+ " [0. 0.0005064 0.02769857 ... 0.02174716 0. 0. ]\n",
+ " [0. 0.00044683 0.02735913 ... 0.02164159 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.06918806250538334, 0.07358180451359603, 0.06781956054169556, 0.08138220770158854, 0.08371857760754807, 0.08447024668642134, 0.0586843891599534, 0.053782920767819035, 0.06942658891697374, 0.04251759111983908] mean of list_RMSE_SousModele 0.06845719495208182\n",
+ " RMSE resultat vote 0.04648178257752944\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 1 value[6] 0.10497512198316751\n",
+ "key 3 value[6] 0.1288294285628787\n",
+ "key 4 value[6] 0.1444116730248319\n",
+ "key 5 value[6] 0.16334850821192484\n",
+ "key 6 value[6] 0.18670485439918758\n",
+ "key 7 value[6] 0.21615527050978905\n",
+ "key 8 value[6] 0.25465259711919624\n",
+ "key 9 value[6] 0.3081780004228992\n",
+ "key 10 value[6] 0.39107838594488675\n",
+ "key 11 value[6] 0.5474158342212556\n",
+ "key 12 value[6] 1.0\n",
+ "key 1 value[2] 0.040480925727196755\n",
+ "key 3 value[2] 0.07778436841936912\n",
+ "key 4 value[2] 0.11952992787819737\n",
+ "key 5 value[2] 0.16592499576261233\n",
+ "key 6 value[2] 0.21916525209290913\n",
+ "key 7 value[2] 0.28110956026359996\n",
+ "key 8 value[2] 0.3558023477795421\n",
+ "key 9 value[2] 0.44405393769607976\n",
+ "key 10 value[2] 0.5581644273121199\n",
+ "key 11 value[2] 0.7232789617283978\n",
+ "key 12 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06342995866787898\n",
+ "Current Error err_H 0.04648178257752944\n",
+ "myFeeder.t 12\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02980312 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00053619 0.0299389 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03007468 ... 0.01879124 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.01432451 ... 0.01778833 0. 0. ]\n",
+ " [0. 0.00044683 0.01432451 ... 0.01810504 0. 0. ]\n",
+ " [0. 0.00044683 0.01432451 ... 0.01810504 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00053619 0.01456212 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00049151 0.01181263 ... 0.01860649 0. 0. ]\n",
+ " [0. 0.00049151 0.00936864 ... 0.01852732 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00384272 0.01323829 ... 0.01821061 0. 0.02440318]\n",
+ " [0. 0.00044683 0.00855397 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00044683 0.00835031 ... 0.01860649 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.0221317 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.02118126 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.0221317 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.01534284 ... 0.01884402 0. 0. ]\n",
+ " [0. 0.00044683 0.01534284 ... 0.01863288 0. 0. ]\n",
+ " [0. 0.00044683 0.01534284 ... 0.01863288 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02566191 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0005064 0.02613714 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02620502 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02620502 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02593347 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.02620502 ... 0.01910794 0. 0. ]\n",
+ " [0. 0.00053619 0.02627291 ... 0.01910794 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.02389681 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00047662 0.02403259 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00047662 0.02403259 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02647658 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02593347 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02579769 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02681602 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.0005064 0.02715547 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.0005064 0.02715547 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03441955 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.03856076 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03767821 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03441955 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.00047662 0.03170401 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.03170401 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00089366 0.02953157 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03156823 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.0305499 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02776646 ... 0.02132489 0. 0. ]\n",
+ " [0. 0.00047662 0.02858113 ... 0.02069148 0. 0. ]\n",
+ " [0. 0.00053619 0.02946368 ... 0.01995249 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03944331 ... 0.01947743 0. 0. ]\n",
+ " [0. 0.00044683 0.03781399 ... 0.02042755 0. 0. ]\n",
+ " [0. 0.00044683 0.03781399 ... 0.02042755 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01170688 0.03279022 ... 0.01826339 0. 0. ]\n",
+ " [0. 0.00047662 0.03265445 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.0010426 0.03156823 ... 0.01852732 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.03496266 ... 0.02074426 0. 0. ]\n",
+ " [0. 0.00047662 0.03374067 ... 0.01963579 0. 0. ]\n",
+ " [0. 0.00044683 0.03380855 ... 0.01963579 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.07238253823210726, 0.08044046261000705, 0.07439543272887654, 0.08970845174679887, 0.08361089277326406, 0.059771447819728936, 0.05584771840387917, 0.08178335594195892, 0.04709025319291115, 0.05072263710658718] mean of list_RMSE_SousModele 0.06957531905561191\n",
+ " RMSE resultat vote 0.04672157710719657\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 1 value[6] 0.0957523625178593\n",
+ "key 3 value[6] 0.11586412658816056\n",
+ "key 4 value[6] 0.1288294285628787\n",
+ "key 5 value[6] 0.1444116730248319\n",
+ "key 6 value[6] 0.16334850821192484\n",
+ "key 8 value[6] 0.21615527050978905\n",
+ "key 9 value[6] 0.25465259711919624\n",
+ "key 10 value[6] 0.3081780004228992\n",
+ "key 11 value[6] 0.39107838594488675\n",
+ "key 12 value[6] 0.5474158342212556\n",
+ "key 13 value[6] 1.0\n",
+ "key 1 value[2] 0.04045420569044003\n",
+ "key 3 value[2] 0.07781600492066869\n",
+ "key 4 value[2] 0.11973099162190343\n",
+ "key 5 value[2] 0.1663733592162449\n",
+ "key 6 value[2] 0.21988486732551982\n",
+ "key 8 value[2] 0.2818517290780968\n",
+ "key 9 value[2] 0.353718201651485\n",
+ "key 10 value[2] 0.4419345336592462\n",
+ "key 11 value[2] 0.5595200445025088\n",
+ "key 12 value[2] 0.7198226198072475\n",
+ "key 13 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06214469854782649\n",
+ "Current Error err_H 0.04672157710719657\n",
+ "myFeeder.t 13\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.01432451 ... 0.01810504 0. 0. ]\n",
+ " [0. 0.00047662 0.0143924 ... 0.01741884 0. 0. ]\n",
+ " [0. 0.00044683 0.0143924 ... 0.01762998 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.16273458 0.11792261 ... 0.03726577 0. 0.27214854]\n",
+ " [0. 0.17551385 0.09891378 ... 0.04228029 0. 0.25800177]\n",
+ " [0. 0.21355377 0.10339443 ... 0.03177619 0. 0.24615385]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.00835031 ... 0.01836896 0. 0. ]\n",
+ " [0. 0.00384272 0.01344196 ... 0.0178939 0. 0.02413793]\n",
+ " [0. 0.00062556 0.00773931 ... 0.01836896 0. 0. ]\n",
+ " ...\n",
+ " [0.14423042 0.14798928 0.06690428 ... 0.02501979 0. 0.21511936]\n",
+ " [0.04929371 0.0792672 0.06395112 ... 0.02351544 0.02031448 0.15649867]\n",
+ " [0.42685746 0.18016086 0.07922607 ... 0.02478226 0. 0.26816976]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.01534284 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00056598 0.01527495 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01527495 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.08641644 0.0976239 ... 0.02866192 0. 0.27480106]\n",
+ " [0.1120406 0.21569854 0.07909029 ... 0.02560042 0. 0.2974359 ]\n",
+ " [0.09466153 0.18439083 0.06476578 ... 0.02396411 0. 0.2627763 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02403259 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.0005064 0.02410048 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00047662 0.02403259 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0.0092827 0.11989872 0.100611 ... 0.02385854 0.00156636 0. ]\n",
+ " [0.04914083 0.15117665 0.06143924 ... 0.02538928 0.01168745 0.08912467]\n",
+ " [0.01433376 0.12406911 0.09898167 ... 0.02977039 0.0016266 0.02847038]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02729124 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00047662 0.02749491 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.0005064 0.02749491 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0.40331438 0.26085791 0.14725051 ... 0.03483769 0. 0.29195402]\n",
+ " [0.34847429 0.25591302 0.13699932 ... 0.0347849 0. 0.29301503]\n",
+ " [0.21426038 0.20390229 0.12715547 ... 0.03161784 0.00684379 0.20353669]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03306178 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.03801765 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00047662 0.03394433 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0.37361952 0.21471552 0.08893415 ... 0.03752969 0. 0.16145004]\n",
+ " [0.38454106 0.28179923 0.15539715 ... 0.04914225 0. 0.22015915]\n",
+ " [0.37015838 0.24000596 0.11276307 ... 0.04069675 0. 0.18744474]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02953157 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.31680071 0.20162933 ... 0.03483769 0. 0.36604775]\n",
+ " [0. 0.31635389 0.18737271 ... 0.03483769 0. 0.35543767]\n",
+ " [0. 0.27613941 0.19450102 ... 0.03483769 0. 0.31299735]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03781399 ... 0.02042755 0. 0. ]\n",
+ " [0. 0.00044683 0.03883232 ... 0.02116653 0. 0. ]\n",
+ " [0. 0.00044683 0.03761032 ... 0.02053312 0. 0. ]\n",
+ " ...\n",
+ " [0.38445545 0.14319333 0.10312288 ... 0.02850356 0. 0.11900973]\n",
+ " [0.39581728 0.19544236 0.10712831 ... 0.03631565 0. 0.20937224]\n",
+ " [0.35703541 0.21105153 0.09633401 ... 0.03626287 0. 0.23076923]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03380855 ... 0.01905516 0. 0. ]\n",
+ " [0. 0.00047662 0.02878479 ... 0.0185801 0. 0. ]\n",
+ " [0. 0.0005064 0.02919212 ... 0.0185801 0. 0. ]\n",
+ " ...\n",
+ " [0.38394178 0.23595472 0.13944331 ... 0.03087886 0. 0.18514589]\n",
+ " [0.37416988 0.20625559 0.10278344 ... 0.03341251 0. 0.24084881]\n",
+ " [0.37501376 0.20348525 0.09904956 ... 0.03267353 0. 0.19964633]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03971487 ... 0.0219583 0. 0. ]\n",
+ " [0. 0.00044683 0.03971487 ... 0.0185801 0. 0. ]\n",
+ " [0. 0.00044683 0.03964698 ... 0.0202692 0. 0. ]\n",
+ " ...\n",
+ " [0.37394973 0.25627048 0.14344874 ... 0.04375825 0. 0.12891247]\n",
+ " [0.37079435 0.24626154 0.13611677 ... 0.03890208 0. 0.14871795]\n",
+ " [0.36120589 0.20759607 0.09979633 ... 0.02966482 0. 0.08152078]]\n",
+ "list_RMSE_SousModele [0.07379127204976892, 0.08023526484202456, 0.0755476237174609, 0.08515955688972586, 0.0637373319067593, 0.06594645397286751, 0.09628506483117148, 0.05733730351013871, 0.056411111334257184, 0.04842025645511599] mean of list_RMSE_SousModele 0.07028712395092902\n",
+ " RMSE resultat vote 0.04980458195514284\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "key 1 value[6] 0.08787852233288299\n",
+ "key 3 value[6] 0.10497512198316751\n",
+ "key 4 value[6] 0.11586412658816056\n",
+ "key 6 value[6] 0.1444116730248319\n",
+ "key 8 value[6] 0.18670485439918758\n",
+ "key 9 value[6] 0.21615527050978905\n",
+ "key 10 value[6] 0.25465259711919624\n",
+ "key 11 value[6] 0.3081780004228992\n",
+ "key 12 value[6] 0.39107838594488675\n",
+ "key 13 value[6] 0.5474158342212556\n",
+ "key 14 value[6] 1.0\n",
+ "key 1 value[2] 0.039712754930650206\n",
+ "key 3 value[2] 0.0762333591638449\n",
+ "key 4 value[2] 0.11715400423475802\n",
+ "key 6 value[2] 0.1627073720193357\n",
+ "key 8 value[2] 0.21484992764542102\n",
+ "key 9 value[2] 0.274666050626177\n",
+ "key 10 value[2] 0.34650170112666895\n",
+ "key 11 value[2] 0.4371164501510711\n",
+ "key 12 value[2] 0.5508698658300257\n",
+ "key 13 value[2] 0.7140604858834921\n",
+ "key 14 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06126326164834909\n",
+ "Current Error err_H 0.04980458195514284\n",
+ "myFeeder.t 14\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657, 0.04980458195514284]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.1792672 0.08180584 ... 0.03288467 0. 0.20424403]\n",
+ " [0. 0.1795353 0.07671419 ... 0.02486144 0. 0.19787798]\n",
+ " [0. 0.19088472 0.07053632 ... 0.0260227 0. 0.24350133]\n",
+ " ...\n",
+ " [0. 0.00047662 0.02878479 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.01663272 ... 0.0185801 0. 0. ]\n",
+ " [0. 0.0005064 0.02797013 ... 0.01910794 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[2.12383049e-01 1.55093834e-01 6.06924644e-02 ... 2.35154394e-02\n",
+ " 0.00000000e+00 2.36870027e-01]\n",
+ " [4.61364887e-01 9.83467382e-02 5.36659878e-02 ... 2.23277910e-02\n",
+ " 0.00000000e+00 1.43766578e-01]\n",
+ " [3.62759127e-01 2.11081323e-01 1.32484725e-01 ... 2.64449723e-02\n",
+ " 7.81131393e-02 2.97612732e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.91510277e-04 7.43380855e-03 ... 1.88440222e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 9.26680244e-03 ... 1.78939034e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 5.09164969e-03 ... 1.89231987e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.14298294 0.13669943 0.07073999 ... 0.02528372 0. 0.24314766]\n",
+ " [0.33104629 0.17664581 0.06116768 ... 0.02660333 0. 0.22475685]\n",
+ " [0.33102183 0.17363718 0.05308893 ... 0.02728952 0. 0.21892131]\n",
+ " ...\n",
+ " [0. 0.00047662 0.01547862 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.01758316 ... 0.01826339 0. 0. ]\n",
+ " [0. 0.00047662 0.02002716 ... 0.02048034 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.04890846 0.13035448 0.11249151 ... 0.03056215 0. 0.03837312]\n",
+ " [0.30791904 0.18439083 0.08913781 ... 0.03399314 0. 0.17206012]\n",
+ " [0.30580322 0.19136133 0.08961303 ... 0.03462655 0. 0.1510168 ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.0287169 ... 0.01953022 0. 0. ]\n",
+ " [0. 0.00047662 0.02613714 ... 0.01942465 0. 0. ]\n",
+ " [0. 0.00044683 0.0287169 ... 0.02042755 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.23384089 0.21843908 0.11934827 ... 0.03093164 0.00087957 0.20265252]\n",
+ " [0.37069651 0.25320226 0.09714868 ... 0.03404592 0. 0.27108753]\n",
+ " [0.3693145 0.23529937 0.07325187 ... 0.03388757 0. 0.26419098]\n",
+ " ...\n",
+ " [0. 0.0005064 0.02443992 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.02892057 ... 0.0185801 0. 0. ]\n",
+ " [0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.37218859 0.20122133 0.0868296 ... 0.03811032 0. 0.18762157]\n",
+ " [0.33297866 0.14402741 0.07583164 ... 0.03457377 0. 0.11653404]\n",
+ " [0.39218492 0.08585046 0.06748133 ... 0.03541832 0. 0.05888594]\n",
+ " ...\n",
+ " [0. 0.00065535 0.03672777 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00062556 0.03625255 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00065535 0.03672777 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.24621782 0.07366696 0.05770536 ... 0.02079704 0. 0.01573828]\n",
+ " [0.09395218 0.05677688 0.05790903 ... 0.02512536 0. 0.11016799]\n",
+ " [0.09395218 0.08388442 0.05947047 ... 0.03035102 0. 0.21644562]\n",
+ " ...\n",
+ " [0. 0.00047662 0.03095723 ... 0.02449195 0. 0. ]\n",
+ " [0. 0.00044683 0.03482688 ... 0.02259171 0. 0. ]\n",
+ " [0. 0.00044683 0.032315 ... 0.02259171 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.3555311 0.20339589 0.11541073 ... 0.03008709 0. 0.13846154]\n",
+ " [0.30051978 0.20548108 0.12260692 ... 0.02924254 0. 0.12961981]\n",
+ " [0.30981471 0.24331248 0.14399185 ... 0.03093164 0. 0.16463307]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03217923 ... 0.01989971 0. 0. ]\n",
+ " [0. 0.00047662 0.03319756 ... 0.01815783 0. 0. ]\n",
+ " [0. 0.00047662 0.02946368 ... 0.01947743 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.35920015 0.21483467 0.099389 ... 0.03045658 0. 0.10910698]\n",
+ " [0.43691066 0.29416145 0.20115411 ... 0.03283188 0. 0.31299735]\n",
+ " [0.43192075 0.30598749 0.20183299 ... 0.03394035 0. 0.33138815]\n",
+ " ...\n",
+ " [0. 0.00044683 0.04229464 ... 0.01910794 0. 0. ]\n",
+ " [0. 0.00044683 0.04446707 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00044683 0.04222675 ... 0.01947743 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.34182719 0.25384272 0.13136456 ... 0.02969121 0. 0.13129973]\n",
+ " [0.37387635 0.28462913 0.14786151 ... 0.046635 0.00856678 0.15649867]\n",
+ " [0.42663731 0.31099196 0.19032587 ... 0.03135392 0.00905476 0.28461538]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03991853 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.0398167 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00049151 0.04114053 ... 0.01900238 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.07357429919010762, 0.08189742768599605, 0.07774387334137628, 0.085483723536597, 0.06850647232554849, 0.06642050935816374, 0.05623775422907881, 0.05673972122092969, 0.04866654506274673, 0.05193456685307547] mean of list_RMSE_SousModele 0.06672048928036199\n",
+ " RMSE resultat vote 0.04642246206846767\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "key 1 value[6] 0.08110499228237947\n",
+ "key 3 value[6] 0.0957523625178593\n",
+ "key 4 value[6] 0.10497512198316751\n",
+ "key 6 value[6] 0.1288294285628787\n",
+ "key 8 value[6] 0.16334850821192484\n",
+ "key 9 value[6] 0.18670485439918758\n",
+ "key 11 value[6] 0.25465259711919624\n",
+ "key 12 value[6] 0.3081780004228992\n",
+ "key 13 value[6] 0.39107838594488675\n",
+ "key 14 value[6] 0.5474158342212556\n",
+ "key 15 value[6] 1.0\n",
+ "key 1 value[2] 0.03917826577907663\n",
+ "key 3 value[2] 0.07547431436330103\n",
+ "key 4 value[2] 0.11609409282863985\n",
+ "key 6 value[2] 0.1613126963193482\n",
+ "key 8 value[2] 0.2130398227481955\n",
+ "key 9 value[2] 0.2726036352780814\n",
+ "key 11 value[2] 0.3438403082798894\n",
+ "key 12 value[2] 0.43082087572243105\n",
+ "key 13 value[2] 0.5442221153441239\n",
+ "key 14 value[2] 0.7050156750515598\n",
+ "key 15 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06027387500969032\n",
+ "Current Error err_H 0.04642246206846767\n",
+ "myFeeder.t 15\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657, 0.04980458195514284, 0.04642246206846767]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02973523 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.00047662 0.02837746 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.0005064 0.0293279 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.00519348 ... 0.01741884 0. 0. ]\n",
+ " [0. 0.00049151 0.00509165 ... 0.01836896 0. 0. ]\n",
+ " [0. 0.00049151 0.00529532 ... 0.01828979 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.01702413 0.00610998 ... 0.02375297 0. 0.1469496 ]\n",
+ " [0. 0.01702413 0.00610998 ... 0.02375297 0. 0.1469496 ]\n",
+ " [0. 0.01693476 0.00610998 ... 0.02375297 0. 0.14668435]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.06404528e-04 2.03665988e-02 ... 2.00580628e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.72437203e-02 ... 1.94246503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 2.38308013e-04 1.09979633e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [0.00000000e+00 1.79773607e-01 3.21792261e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 2.63483643e-01]\n",
+ " [0.00000000e+00 1.79773607e-01 3.21792261e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 2.63483643e-01]\n",
+ " [0.00000000e+00 1.79773607e-01 3.21792261e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 2.63483643e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02600136 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00047662 0.02600136 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02443992 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.15874293 0.02701969 ... 0.02507258 0. 0.1193634 ]\n",
+ " [0. 0.15874293 0.02701969 ... 0.02507258 0. 0.1193634 ]\n",
+ " [0. 0.15874293 0.02701969 ... 0.02507258 0. 0.1193634 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00065535 0.03632043 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00062556 0.03672777 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00065535 0.03672777 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.22195413 0.04161575 ... 0.01826339 0. 0.17046861]\n",
+ " [0. 0.22195413 0.04161575 ... 0.01826339 0. 0.17046861]\n",
+ " [0. 0.22195413 0.04161575 ... 0.01826339 0. 0.17046861]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03095723 ... 0.02216944 0. 0. ]\n",
+ " [0. 0.00044683 0.0371351 ... 0.02079704 0. 0. ]\n",
+ " [0. 0.00044683 0.0311609 ... 0.02438638 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.0311609 ... 0.01947743 0. 0. ]\n",
+ " [0. 0.00044683 0.03224711 ... 0.01984693 0. 0. ]\n",
+ " [0. 0.00047662 0.03197556 ... 0.0192663 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.17566279 0.06367957 ... 0.02459752 0. 0.13616269]\n",
+ " [0. 0.17566279 0.06367957 ... 0.02459752 0. 0.13616269]\n",
+ " [0. 0.17566279 0.06367957 ... 0.02459752 0. 0.13616269]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03937542 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03971487 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00044683 0.04426341 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]\n",
+ " [0. 0.16753053 0.04467074 ... 0.02623383 0. 0.13757737]\n",
+ " [0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.04205703 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00044683 0.04063136 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00044683 0.03767821 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00308311 0.04124236 ... 0.01979414 0. 0.0331565 ]\n",
+ " [0. 0.00308311 0.04124236 ... 0.01979414 0. 0.0331565 ]\n",
+ " [0. 0.00308311 0.04124236 ... 0.01979414 0. 0.0331565 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03767821 ... 0.02216944 0. 0. ]\n",
+ " [0. 0.00044683 0.04582485 ... 0.01979414 0. 0. ]\n",
+ " [0. 0.00044683 0.03767821 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.01653262 0.0407332 ... 0.0253365 0. 0.35543767]\n",
+ " [0. 0.01653262 0.0407332 ... 0.0253365 0. 0.35543767]\n",
+ " [0. 0.01653262 0.0407332 ... 0.02612827 0. 0.32625995]]\n",
+ "list_RMSE_SousModele [0.06513506088352221, 0.0731398573715648, 0.06674005824762508, 0.06050164770232898, 0.06288690912556744, 0.053462892877013175, 0.053254085907502975, 0.048219742888484324, 0.043933191578461164, 0.0550362666415106] mean of list_RMSE_SousModele 0.058230971322358084\n",
+ " RMSE resultat vote 0.04025907547983262\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "key 1 value[6] 0.0752353553817643\n",
+ "key 3 value[6] 0.08787852233288299\n",
+ "key 4 value[6] 0.0957523625178593\n",
+ "key 8 value[6] 0.1444116730248319\n",
+ "key 9 value[6] 0.16334850821192484\n",
+ "key 11 value[6] 0.21615527050978905\n",
+ "key 12 value[6] 0.25465259711919624\n",
+ "key 13 value[6] 0.3081780004228992\n",
+ "key 14 value[6] 0.39107838594488675\n",
+ "key 15 value[6] 0.5474158342212556\n",
+ "key 16 value[6] 1.0\n",
+ "key 1 value[2] 0.0393189205157291\n",
+ "key 3 value[2] 0.0758294580060259\n",
+ "key 4 value[2] 0.11653328782500504\n",
+ "key 8 value[2] 0.16190210264149102\n",
+ "key 9 value[2] 0.21362338111180387\n",
+ "key 11 value[2] 0.2740574412671437\n",
+ "key 12 value[2] 0.3451852463081013\n",
+ "key 13 value[2] 0.43281102010587924\n",
+ "key 14 value[2] 0.5461000755149809\n",
+ "key 15 value[2] 0.7074078087271731\n",
+ "key 16 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.05902295003907423\n",
+ "Current Error err_H 0.04025907547983262\n",
+ "myFeeder.t 16\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657, 0.04980458195514284, 0.04642246206846767, 0.04025907547983262]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " ...\n",
+ " [0.40069712 0.35963658 0.31228785 ... 0.02897862 0.14558708 0.49425287]\n",
+ " [0.40069712 0.35963658 0.31228785 ... 0.02897862 0.14558708 0.49425287]\n",
+ " [0.40069712 0.35963658 0.31228785 ... 0.02897862 0.14558708 0.49425287]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.17977361 0.03217923 ... 0.01821061 0. 0.26348364]\n",
+ " [0. 0.17977361 0.03217923 ... 0.01821061 0. 0.26348364]\n",
+ " [0. 0.17977361 0.03217923 ... 0.01821061 0. 0.26348364]\n",
+ " ...\n",
+ " [0.47423714 0.36207924 0.3389002 ... 0.03145949 0.19922887 0.488771 ]\n",
+ " [0.47423714 0.36207924 0.3389002 ... 0.03145949 0.19922887 0.488771 ]\n",
+ " [0.47423714 0.36207924 0.3389002 ... 0.03145949 0.19922887 0.488771 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.15853441 0.02708758 ... 0.02507258 0. 0.11794872]\n",
+ " [0. 0.15853441 0.0269518 ... 0.02501979 0. 0.11794872]\n",
+ " [0. 0.15823652 0.0269518 ... 0.02491423 0. 0.11653404]\n",
+ " ...\n",
+ " [0.44209625 0.369437 0.33340122 ... 0.04206915 0.16682933 0.44686118]\n",
+ " [0.44209625 0.369437 0.33340122 ... 0.04206915 0.16682933 0.44686118]\n",
+ " [0.46375589 0.37003277 0.33021045 ... 0.03531275 0.17251642 0.44739169]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.22195413 0.04161575 ... 0.01826339 0. 0.17046861]\n",
+ " [0. 0.22272863 0.04154786 ... 0.01826339 0. 0.17082228]\n",
+ " [0. 0.22237116 0.04120842 ... 0.01826339 0. 0.16870027]\n",
+ " ...\n",
+ " [0.35853972 0.36279416 0.36619145 ... 0.0395355 0.09271643 0.44420866]\n",
+ " [0.35853972 0.36279416 0.36619145 ... 0.0395355 0.09271643 0.44420866]\n",
+ " [0.35853972 0.36279416 0.36619145 ... 0.0395355 0.09271643 0.44420866]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " ...\n",
+ " [0.38188712 0.33848674 0.34120842 ... 0.03462655 0.12169408 0.41061008]\n",
+ " [0.38188712 0.33848674 0.34120842 ... 0.03462655 0.12169408 0.41061008]\n",
+ " [0.38425977 0.34206136 0.34405974 ... 0.03462655 0.12165793 0.41662246]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.17566279 0.06367957 ... 0.02459752 0. 0.13616269]\n",
+ " [0. 0.17566279 0.06367957 ... 0.02459752 0. 0.13616269]\n",
+ " [0. 0.17566279 0.06367957 ... 0.02459752 0. 0.13616269]\n",
+ " ...\n",
+ " [0.39386045 0.3468871 0.34392396 ... 0.04022169 0.17619134 0.43908046]\n",
+ " [0.39480218 0.34721478 0.34677529 ... 0.03510161 0.17586602 0.43819629]\n",
+ " [0.4042561 0.34721478 0.34718262 ... 0.04022169 0.18248087 0.44314766]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]\n",
+ " [0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]\n",
+ " [0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]\n",
+ " ...\n",
+ " [0.37485477 0.34864462 0.34453496 ... 0.03177619 0.1281282 0.47975243]\n",
+ " [0.37485477 0.34864462 0.34453496 ... 0.03177619 0.1281282 0.47975243]\n",
+ " [0.37392527 0.3484361 0.34494229 ... 0.03198733 0.13439364 0.4795756 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00308311 0.04124236 ... 0.01979414 0. 0.0331565 ]\n",
+ " [0. 0.00567471 0.04083503 ... 0.01995249 0. 0.066313 ]\n",
+ " [0. 0.00567471 0.04083503 ... 0.01995249 0. 0.066313 ]\n",
+ " ...\n",
+ " [0.36837278 0.35142985 0.34837067 ... 0.053365 0.12824869 0.4734748 ]\n",
+ " [0.38736012 0.35062556 0.34847251 ... 0.03483769 0.14446051 0.47214854]\n",
+ " [0.38756192 0.35062556 0.35010183 ... 0.03483769 0.14070125 0.47214854]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01653262 0.0407332 ... 0.02612827 0. 0.32625995]\n",
+ " [0. 0.01653262 0.0407332 ... 0.02612827 0. 0.32625995]\n",
+ " [0. 0.01653262 0.0407332 ... 0.02612827 0. 0.32625995]\n",
+ " ...\n",
+ " [0.39735828 0.35567471 0.33910387 ... 0.03167063 0.18561359 0.52254642]\n",
+ " [0.39735828 0.35567471 0.33910387 ... 0.03167063 0.18561359 0.52254642]\n",
+ " [0.39735828 0.35567471 0.33910387 ... 0.03167063 0.18561359 0.52254642]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 3.05498982e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.05498982e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.05498982e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [3.68739681e-01 3.41823056e-01 3.15682281e-01 ... 3.00870942e-02\n",
+ " 1.79830110e-01 5.41114058e-01]\n",
+ " [3.68739681e-01 3.41823056e-01 3.15682281e-01 ... 3.00870942e-02\n",
+ " 1.79830110e-01 5.41114058e-01]\n",
+ " [3.68739681e-01 3.41823056e-01 3.15682281e-01 ... 3.00870942e-02\n",
+ " 1.79830110e-01 5.41114058e-01]]\n",
+ "list_RMSE_SousModele [0.07054819272510317, 0.0674892813952671, 0.07085544345128844, 0.09214879742610274, 0.07580005152596367, 0.07374958029313652, 0.07727621737184309, 0.07743845301119924, 0.08609947314173202, 0.08047079030629137] mean of list_RMSE_SousModele 0.07718762806479273\n",
+ " RMSE resultat vote 0.06302297235310445\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 1 value[6] 0.0701133935324861\n",
+ "key 4 value[6] 0.08787852233288299\n",
+ "key 8 value[6] 0.1288294285628787\n",
+ "key 9 value[6] 0.1444116730248319\n",
+ "key 11 value[6] 0.18670485439918758\n",
+ "key 12 value[6] 0.21615527050978905\n",
+ "key 13 value[6] 0.25465259711919624\n",
+ "key 14 value[6] 0.3081780004228992\n",
+ "key 15 value[6] 0.39107838594488675\n",
+ "key 16 value[6] 0.5474158342212556\n",
+ "key 17 value[6] 1.0\n",
+ "key 1 value[2] 0.04052551288297275\n",
+ "key 4 value[2] 0.07678745182438725\n",
+ "key 8 value[2] 0.11677564759411514\n",
+ "key 9 value[2] 0.16186047666175396\n",
+ "key 11 value[2] 0.2142356900514927\n",
+ "key 12 value[2] 0.27415556385183665\n",
+ "key 13 value[2] 0.34485013441644663\n",
+ "key 14 value[2] 0.43243428208561163\n",
+ "key 15 value[2] 0.5469735161513577\n",
+ "key 16 value[2] 0.7103617103502106\n",
+ "key 17 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.0592582454693113\n",
+ "Current Error err_H 0.06302297235310445\n",
+ "myFeeder.t 17\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657, 0.04980458195514284, 0.04642246206846767, 0.04025907547983262, 0.06302297235310445]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.72818443e-01 3.55942806e-01 2.98031229e-01 ... 2.86619161e-02\n",
+ " 1.57141997e-01 4.98320071e-01]\n",
+ " [4.78040726e-01 3.53410783e-01 2.84657162e-01 ... 2.85563473e-02\n",
+ " 1.56744382e-01 5.00442087e-01]\n",
+ " [4.74188222e-01 3.54959786e-01 2.94976239e-01 ... 2.86619161e-02\n",
+ " 1.57141997e-01 4.98320071e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 1.64969450e-02 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 1.64969450e-02 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 1.83299389e-02 ... 1.84745315e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.05631994e-01 3.61274948e-01 3.41887305e-01 ... 3.14594880e-02\n",
+ " 1.98915597e-01 4.91069850e-01]\n",
+ " [5.60276402e-01 3.63240989e-01 3.43720299e-01 ... 3.04038005e-02\n",
+ " 1.98710766e-01 5.00972591e-01]\n",
+ " [6.29376873e-01 3.61930295e-01 3.48472505e-01 ... 3.08260755e-02\n",
+ " 1.97746852e-01 5.01503095e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 1.55465037e-02 ... 1.87912378e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.54107264e-02 ... 1.87912378e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.55465037e-02 ... 1.87912378e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.63645814e-01 3.70539172e-01 3.31228785e-01 ... 3.53127474e-02\n",
+ " 1.72697150e-01 4.48099027e-01]\n",
+ " [4.80535682e-01 3.65564492e-01 3.27359131e-01 ... 3.56294537e-02\n",
+ " 1.72829689e-01 4.52343059e-01]\n",
+ " [5.37760656e-01 3.66815609e-01 3.27494908e-01 ... 3.58933756e-02\n",
+ " 1.79143322e-01 4.54465075e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.36193029e-04 2.89884589e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.89205703e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.89205703e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.08512200e-01 3.43312481e-01 3.57094365e-01 ... 3.41514912e-02\n",
+ " 1.44454485e-01 4.10610080e-01]\n",
+ " [5.17348499e-01 3.40601728e-01 3.38424983e-01 ... 3.34125099e-02\n",
+ " 1.63805048e-01 4.12908930e-01]\n",
+ " [5.31462117e-01 3.43521001e-01 3.40665309e-01 ... 3.37820005e-02\n",
+ " 1.48816194e-01 4.12378426e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.73387644e-02 ... 2.22750066e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.04616429e-02 ... 2.15888097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.07331976e-02 ... 2.13776722e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.01504311e-01 3.48287161e-01 3.51595384e-01 ... 3.50488255e-02\n",
+ " 1.81709742e-01 4.36427940e-01]\n",
+ " [5.34862105e-01 3.49746798e-01 3.43788187e-01 ... 3.51016099e-02\n",
+ " 1.68419784e-01 4.32183908e-01]\n",
+ " [5.61426038e-01 3.59994042e-01 3.41004752e-01 ... 3.53127474e-02\n",
+ " 1.81974818e-01 4.67550840e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.06404528e-04 3.41479973e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.45553293e-02 ... 1.91079440e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.46911066e-02 ... 1.91079440e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[3.93273405e-01 3.46201966e-01 3.47114732e-01 ... 3.19345474e-02\n",
+ " 1.27489608e-01 4.75685234e-01]\n",
+ " [5.21604599e-01 3.39469765e-01 3.40529532e-01 ... 3.81103193e-02\n",
+ " 1.54792457e-01 4.65959328e-01]\n",
+ " [5.42481502e-01 3.49180816e-01 3.41819416e-01 ... 3.40987068e-02\n",
+ " 1.66829327e-01 4.63306808e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 4.42634080e-02 ... 1.95302191e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.31093007e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.18873048e-02 ... 1.92662972e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.47807742e-01 3.53887399e-01 3.57331976e-01 ... 3.48376880e-02\n",
+ " 1.23531538e-01 4.74005305e-01]\n",
+ " [4.52394056e-01 3.56300268e-01 3.39816701e-01 ... 3.86381631e-02\n",
+ " 1.62823062e-01 4.77453581e-01]\n",
+ " [6.66923500e-01 3.66487936e-01 3.56517312e-01 ... 3.47585115e-02\n",
+ " 1.16898608e-01 4.73740053e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.91038697e-02 ... 1.83689628e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 4.06313646e-02 ... 1.83689628e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.17515275e-02 ... 1.83689628e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[3.87268391e-01 3.57462020e-01 3.29938900e-01 ... 2.69200317e-02\n",
+ " 1.83986987e-01 4.66843501e-01]\n",
+ " [3.95523757e-01 3.48525469e-01 3.31975560e-01 ... 2.77117973e-02\n",
+ " 1.84529189e-01 4.74801061e-01]\n",
+ " [4.86699688e-01 3.20822163e-01 3.16700611e-01 ... 2.77117973e-02\n",
+ " 1.79287909e-01 4.72148541e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 4.07331976e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.86965377e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.07331976e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[3.74977068e-01 3.54781055e-01 3.42158859e-01 ... 2.29612035e-02\n",
+ " 1.81275980e-01 5.09283820e-01]\n",
+ " [3.74977068e-01 3.54781055e-01 3.42158859e-01 ... 2.29612035e-02\n",
+ " 1.81275980e-01 5.09283820e-01]\n",
+ " [5.46505228e-01 3.48972297e-01 3.15682281e-01 ... 2.29612035e-02\n",
+ " 1.83806253e-01 5.01326260e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.66598778e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.66598778e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.66598778e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.71522045e-01 3.43997617e-01 3.48472505e-01 ... 3.59989443e-02\n",
+ " 1.38430026e-01 5.22900088e-01]\n",
+ " [4.84926313e-01 3.38993149e-01 3.45281738e-01 ... 3.08260755e-02\n",
+ " 1.50792216e-01 5.22546419e-01]\n",
+ " [5.50235431e-01 3.44474233e-01 3.40801086e-01 ... 3.17234099e-02\n",
+ " 1.23019459e-01 5.20954907e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 3.15003394e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.33333333e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.65919891e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.0634462955129073, 0.06406274405381332, 0.06179048808901572, 0.06044073063835678, 0.060704174319012205, 0.06106228828206411, 0.060841148289128205, 0.07632430385773724, 0.06970780978365027, 0.04427280264245068] mean of list_RMSE_SousModele 0.062265278546813574\n",
+ " RMSE resultat vote 0.045869223165774825\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 1 value[6] 0.0656141137913348\n",
+ "key 4 value[6] 0.08110499228237947\n",
+ "key 8 value[6] 0.11586412658816056\n",
+ "key 11 value[6] 0.16334850821192484\n",
+ "key 12 value[6] 0.18670485439918758\n",
+ "key 13 value[6] 0.21615527050978905\n",
+ "key 14 value[6] 0.25465259711919624\n",
+ "key 15 value[6] 0.3081780004228992\n",
+ "key 16 value[6] 0.39107838594488675\n",
+ "key 17 value[6] 0.5474158342212556\n",
+ "key 18 value[6] 1.0\n",
+ "key 1 value[2] 0.040760630733471054\n",
+ "key 4 value[2] 0.07794158282040015\n",
+ "key 8 value[2] 0.11914030458282023\n",
+ "key 11 value[2] 0.16529313547795918\n",
+ "key 12 value[2] 0.21798142001742124\n",
+ "key 13 value[2] 0.2788565549934844\n",
+ "key 14 value[2] 0.35083843182889585\n",
+ "key 15 value[2] 0.43971749880657085\n",
+ "key 16 value[2] 0.5558642462378697\n",
+ "key 17 value[2] 0.720335136761988\n",
+ "key 18 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.05851441089689261\n",
+ "Current Error err_H 0.045869223165774825\n",
+ "myFeeder.t 18\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657, 0.04980458195514284, 0.04642246206846767, 0.04025907547983262, 0.06302297235310445, 0.045869223165774825]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.12491514e-02 ... 1.87384534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.08418194e-02 ... 1.84745315e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.17039023e-04 1.70400543e-02 ... 1.84217472e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.86069834e-01 3.51921358e-01 3.47386286e-01 ... 2.78701504e-02\n",
+ " 1.74504488e-01 4.79929266e-01]\n",
+ " [5.86155445e-01 3.52874590e-01 3.53496266e-01 ... 2.79757192e-02\n",
+ " 1.87517320e-01 4.78337754e-01]\n",
+ " [5.84406531e-01 3.52785225e-01 3.52002716e-01 ... 2.83979942e-02\n",
+ " 1.86987168e-01 4.77630416e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.76616026e-04 1.54786151e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.55465037e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.57501697e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [7.27083716e-01 3.25707477e-01 3.04209097e-01 ... 2.99815255e-02\n",
+ " 1.97529972e-01 4.90539346e-01]\n",
+ " [7.26508897e-01 3.29550194e-01 3.09979633e-01 ... 3.00343098e-02\n",
+ " 1.97626363e-01 4.90893015e-01]\n",
+ " [7.29713202e-01 3.26511766e-01 3.09097081e-01 ... 3.44681974e-02\n",
+ " 1.98011928e-01 4.85764810e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02844535 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.02851324 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.0005064 0.02885268 ... 0.01921351 0. 0. ]\n",
+ " ...\n",
+ " [0.43881857 0.3497468 0.34487441 ... 0.03394035 0.093596 0.40212202]\n",
+ " [0.45931633 0.35108728 0.34480652 ... 0.03404592 0.09352371 0.40229885]\n",
+ " [0.47064147 0.34673816 0.3424983 ... 0.03372922 0.0930538 0.40707339]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.0407332 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.00047662 0.03272234 ... 0.02370018 0. 0. ]\n",
+ " [0. 0.00044683 0.03340122 ... 0.02248614 0. 0. ]\n",
+ " ...\n",
+ " [0.18293891 0.19133155 0.18065173 ... 0.03198733 0.06702813 0.31335102]\n",
+ " [0.20964961 0.21027703 0.20074678 ... 0.03225125 0.0796554 0.32484527]\n",
+ " [0.28630832 0.26273458 0.2496945 ... 0.03314859 0.10503042 0.35932803]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.76616026e-04 3.45553293e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 3.36727766e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 3.23828921e-02 ... 1.87912378e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.59970648e-01 3.61334525e-01 3.41208418e-01 ... 3.61572974e-02\n",
+ " 1.57310681e-01 4.28116711e-01]\n",
+ " [5.59970648e-01 3.61334525e-01 3.41208418e-01 ... 3.61572974e-02\n",
+ " 1.57310681e-01 4.28116711e-01]\n",
+ " [5.59970648e-01 3.61334525e-01 3.41208418e-01 ... 3.61572974e-02\n",
+ " 1.57310681e-01 4.28116711e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 3.96469790e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.34487441e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.43991853e-02 ... 1.94774347e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [4.80132086e-01 3.39469765e-01 3.13577733e-01 ... 3.38875693e-02\n",
+ " 1.02644738e-01 4.30769231e-01]\n",
+ " [4.40836544e-01 3.43997617e-01 3.32993890e-01 ... 3.39931380e-02\n",
+ " 1.21151877e-01 4.48629531e-01]\n",
+ " [3.86705803e-01 3.49270182e-01 3.47522064e-01 ... 3.43098443e-02\n",
+ " 1.31140430e-01 4.66666667e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 4.07331976e-02 ... 1.86856690e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 4.08350305e-02 ... 1.86856690e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.23625255e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.88295726e-01 3.54378910e-01 2.96537678e-01 ... 3.45209818e-02\n",
+ " 1.31050063e-01 4.77188329e-01]\n",
+ " [5.78609429e-01 3.41018767e-01 2.70875764e-01 ... 3.31749802e-02\n",
+ " 1.13844207e-01 4.61273210e-01]\n",
+ " [5.43753440e-01 3.60008937e-01 3.09368635e-01 ... 3.43626287e-02\n",
+ " 1.39472257e-01 4.74005305e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03665988 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.04480652 ... 0.01979414 0. 0. ]\n",
+ " [0. 0.00044683 0.04480652 ... 0.01979414 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.0305499 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.06404528e-04 3.71350984e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.79497624e-02 ... 1.91607284e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.65241005e-02 ... 1.92135128e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.36378646e-01 3.59398272e-01 3.32993890e-01 ... 3.07732911e-02\n",
+ " 1.81107296e-01 5.40583554e-01]\n",
+ " [5.36378646e-01 3.59398272e-01 3.32993890e-01 ... 3.07732911e-02\n",
+ " 1.81107296e-01 5.40583554e-01]\n",
+ " [5.36378646e-01 3.59398272e-01 3.32993890e-01 ... 3.07732911e-02\n",
+ " 1.81107296e-01 5.40583554e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.06404528e-04 3.57094365e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.59809912e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.66598778e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.17006054e-01 3.56687519e-01 3.19076714e-01 ... 2.87147004e-02\n",
+ " 1.31140430e-01 5.33863837e-01]\n",
+ " [5.06121201e-01 3.55793864e-01 3.15003394e-01 ... 2.84507786e-02\n",
+ " 1.29152359e-01 5.29089302e-01]\n",
+ " [4.93695346e-01 3.55436402e-01 3.20027155e-01 ... 2.84507786e-02\n",
+ " 1.34212904e-01 5.32272325e-01]]\n",
+ "list_RMSE_SousModele [0.07005809004326741, 0.07190566611499132, 0.06706070228249626, 0.06842351091427931, 0.07013575475417834, 0.07310672912315051, 0.0774712364294957, 0.08068472241820589, 0.05452830753362161, 0.05138410398329418] mean of list_RMSE_SousModele 0.06847588235969805\n",
+ " RMSE resultat vote 0.05168623143618591\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 1 value[6] 0.06163692269875859\n",
+ "key 4 value[6] 0.0752353553817643\n",
+ "key 8 value[6] 0.10497512198316751\n",
+ "key 11 value[6] 0.1444116730248319\n",
+ "key 12 value[6] 0.16334850821192484\n",
+ "key 13 value[6] 0.18670485439918758\n",
+ "key 14 value[6] 0.21615527050978905\n",
+ "key 16 value[6] 0.3081780004228992\n",
+ "key 17 value[6] 0.39107838594488675\n",
+ "key 18 value[6] 0.5474158342212556\n",
+ "key 19 value[6] 1.0\n",
+ "key 1 value[2] 0.03994061997605656\n",
+ "key 4 value[2] 0.07653900723559813\n",
+ "key 8 value[2] 0.11704876293927932\n",
+ "key 11 value[2] 0.16244032484237222\n",
+ "key 12 value[2] 0.2143995800922803\n",
+ "key 13 value[2] 0.2744831109461506\n",
+ "key 14 value[2] 0.3459562318078342\n",
+ "key 16 value[2] 0.43496423761323355\n",
+ "key 17 value[2] 0.5510352729270175\n",
+ "key 18 value[2] 0.7133453026771815\n",
+ "key 19 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.05815503303053963\n",
+ "Current Error err_H 0.05168623143618591\n",
+ "myFeeder.t 19\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657, 0.04980458195514284, 0.04642246206846767, 0.04025907547983262, 0.06302297235310445, 0.045869223165774825, 0.05168623143618591]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.90509387e-01 3.47453083e-01 3.36591989e-01 ... 2.79757192e-02\n",
+ " 1.88529429e-01 4.87002653e-01]\n",
+ " [5.88026662e-01 3.27852249e-01 3.02036660e-01 ... 2.72895223e-02\n",
+ " 1.83709862e-01 4.68435013e-01]\n",
+ " [5.82975601e-01 3.14060173e-01 2.90088255e-01 ... 2.75534442e-02\n",
+ " 1.89288511e-01 4.57648099e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.53903598e-02 ... 1.88440222e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.35573659e-02 ... 1.89495909e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.41887305e-02 ... 1.78939034e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.30288021e-01 3.24039321e-01 3.04820095e-01 ... 3.45209818e-02\n",
+ " 1.98746912e-01 4.73916888e-01]\n",
+ " [5.98055403e-01 3.36788800e-01 3.19076714e-01 ... 3.48376880e-02\n",
+ " 1.96770890e-01 4.34836428e-01]\n",
+ " [5.58490797e-01 3.12779267e-01 2.57026477e-01 ... 3.31485880e-02\n",
+ " 1.69540334e-01 3.91335102e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 1.83299389e-02 ... 2.04803378e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 1.71758316e-02 ... 1.98469253e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.21995927e-02 ... 1.96357878e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.58038280e-01 3.50580876e-01 3.56143924e-01 ... 3.33069411e-02\n",
+ " 9.41863968e-02 4.34305924e-01]\n",
+ " [4.41912799e-01 3.50848972e-01 3.27698574e-01 ... 3.17234099e-02\n",
+ " 9.14874390e-02 4.35897436e-01]\n",
+ " [4.39870360e-01 3.49865952e-01 3.16836388e-01 ... 3.39931380e-02\n",
+ " 1.10693415e-01 4.40848806e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.44399185e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.44399185e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.86833482e-03 2.86490156e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 3.00618921e-03]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.87311197e-01 3.45546619e-01 3.37746096e-01 ... 3.42042755e-02\n",
+ " 1.47454666e-01 4.14854111e-01]\n",
+ " [4.06164007e-01 3.58802502e-01 3.38289206e-01 ... 3.41514912e-02\n",
+ " 1.34321345e-01 4.13085765e-01]\n",
+ " [3.81312297e-01 3.60887697e-01 3.42226748e-01 ... 3.42570599e-02\n",
+ " 1.41888066e-01 4.13969938e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 3.17040054e-02 ... 2.29612035e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.01425662e-02 ... 1.94774347e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.17718941e-02 ... 2.07970441e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.28954932e-01 3.60172773e-01 3.39511202e-01 ... 3.60517287e-02\n",
+ " 1.50960901e-01 4.27232538e-01]\n",
+ " [4.29939461e-01 3.52546917e-01 3.34487441e-01 ... 3.47849037e-02\n",
+ " 1.52659799e-01 4.34659593e-01]\n",
+ " [3.83538189e-01 3.51474531e-01 3.31093007e-01 ... 3.48904724e-02\n",
+ " 1.55611784e-01 4.38903625e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.36193029e-04 3.50305499e-02 ... 1.95302191e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.63883232e-02 ... 1.96357878e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.60488798e-02 ... 1.91607284e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.12902831e-01 3.49210605e-01 3.44738629e-01 ... 3.43098443e-02\n",
+ " 1.32501958e-01 4.67374005e-01]\n",
+ " [3.92747508e-01 3.50431933e-01 3.23964698e-01 ... 3.40459224e-02\n",
+ " 1.14091210e-01 4.67904509e-01]\n",
+ " [4.15030881e-01 3.54810843e-01 3.19212492e-01 ... 3.38347849e-02\n",
+ " 1.13175492e-01 4.64544651e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 4.26340801e-02 ... 1.95830034e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.76103191e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.65919891e-02 ... 1.94246503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.34617501e-01 3.60634495e-01 3.13951120e-01 ... 3.34125099e-02\n",
+ " 1.28284836e-01 4.59151194e-01]\n",
+ " [6.94551458e-01 3.25692583e-01 3.15478615e-01 ... 3.29374505e-02\n",
+ " 1.47460690e-01 4.60477454e-01]\n",
+ " [5.19519354e-01 3.48570152e-01 3.30549898e-01 ... 3.23040380e-02\n",
+ " 1.38116754e-01 4.76657825e-01]\n",
+ " ...\n",
+ " [8.62227114e-03 4.46827525e-04 3.11608961e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.31975560e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [3.63236103e-02 5.80875782e-04 4.40936864e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.40096618e-01 3.56449211e-01 3.31160896e-01 ... 3.08260755e-02\n",
+ " 1.80504850e-01 5.36870027e-01]\n",
+ " [4.93744267e-01 3.44831695e-01 3.35709437e-01 ... 3.81631037e-02\n",
+ " 1.79528887e-01 5.46065429e-01]\n",
+ " [6.59988993e-01 3.30711945e-01 3.36863544e-01 ... 3.58933756e-02\n",
+ " 1.41514549e-01 5.34217507e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.10930075e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.34012220e-02 ... 1.92662972e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.23828921e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.87213355e-01 3.54572535e-01 3.34555329e-01 ... 2.93481130e-02\n",
+ " 1.24007470e-01 5.35809019e-01]\n",
+ " [3.95939583e-01 3.50938338e-01 3.49083503e-01 ... 3.08788599e-02\n",
+ " 1.30537984e-01 5.28205128e-01]\n",
+ " [4.03950346e-01 3.40988978e-01 3.41412084e-01 ... 2.99287411e-02\n",
+ " 1.43695403e-01 5.21662246e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.69314325e-02 ... 1.94246503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.17718941e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.25865580e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.83085672e-01 3.64164433e-01 3.62627291e-01 ... 2.86619161e-02\n",
+ " 1.33200795e-01 5.09283820e-01]\n",
+ " [4.37919648e-01 3.67962466e-01 3.38798371e-01 ... 3.59461599e-02\n",
+ " 1.66744985e-01 5.16976127e-01]\n",
+ " [5.75564117e-01 3.65728329e-01 3.52342159e-01 ... 2.77117973e-02\n",
+ " 1.06036508e-01 5.02652520e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.23828921e-02 ... 2.12193191e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 6.07685433e-03 3.22810591e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 2.81501340e-03 3.22810591e-02 ... 1.97941409e-02\n",
+ " 0.00000000e+00 4.24403183e-02]]\n",
+ "list_RMSE_SousModele [0.07463769895332431, 0.08922144474761692, 0.0695089934020948, 0.06678827946261061, 0.07811915912376366, 0.070212288143011, 0.08834657851073231, 0.07220322517914862, 0.0727082393569627, 0.0623236653130574] mean of list_RMSE_SousModele 0.07440695721923223\n",
+ " RMSE resultat vote 0.05779109135515785\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "key 1 value[6] 0.058100387712334015\n",
+ "key 4 value[6] 0.0701133935324861\n",
+ "key 8 value[6] 0.0957523625178593\n",
+ "key 11 value[6] 0.1288294285628787\n",
+ "key 12 value[6] 0.1444116730248319\n",
+ "key 13 value[6] 0.16334850821192484\n",
+ "key 14 value[6] 0.18670485439918758\n",
+ "key 17 value[6] 0.3081780004228992\n",
+ "key 18 value[6] 0.39107838594488675\n",
+ "key 19 value[6] 0.5474158342212556\n",
+ "key 20 value[6] 1.0\n",
+ "key 1 value[2] 0.04058119616689651\n",
+ "key 4 value[2] 0.07734088883613789\n",
+ "key 8 value[2] 0.1181604985212059\n",
+ "key 11 value[2] 0.1634605336522444\n",
+ "key 12 value[2] 0.21497699551990407\n",
+ "key 13 value[2] 0.2745036491251996\n",
+ "key 14 value[2] 0.34436130249087105\n",
+ "key 17 value[2] 0.4316441160195559\n",
+ "key 18 value[2] 0.5459592623348042\n",
+ "key 19 value[2] 0.7086077043738703\n",
+ "key 20 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.058136835946770535\n",
+ "Current Error err_H 0.05779109135515785\n",
+ "myFeeder.t 20\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657, 0.04980458195514284, 0.04642246206846767, 0.04025907547983262, 0.06302297235310445, 0.045869223165774825, 0.05168623143618591, 0.05779109135515785]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.01473184 ... 0.01773555 0. 0. ]\n",
+ " [0. 0.00223414 0.0191446 ... 0.01794669 0. 0. ]\n",
+ " [0. 0.00226393 0.02043449 ... 0.01815783 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.0143924 ... 0.0178939 0. 0. ]\n",
+ " [0. 0.00044683 0.01446029 ... 0.01794669 0. 0. ]\n",
+ " [0. 0.00047662 0.01432451 ... 0.01752441 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00053619 0.02552614 ... 0.01889681 0. 0.00194518]\n",
+ " [0. 0.00563003 0.02885268 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.01283884 0.03306178 ... 0.01916073 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00988978 0.02715547 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00988978 0.02715547 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00616622 0.02729124 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02864902 ... 0.02306677 0. 0. ]\n",
+ " [0. 0.00044683 0.03048201 ... 0.02185273 0. 0. ]\n",
+ " [0. 0.00047662 0.034759 ... 0.02164159 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03957909 ... 0.02158881 0. 0. ]\n",
+ " [0. 0.00044683 0.03957909 ... 0.02158881 0. 0. ]\n",
+ " [0. 0.00044683 0.03957909 ... 0.02153603 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03408011 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00047662 0.03136456 ... 0.0192663 0. 0. ]\n",
+ " [0. 0.0005064 0.03312967 ... 0.01916073 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03611677 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00044683 0.03611677 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00044683 0.03570944 ... 0.01879124 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03598099 ... 0.01942465 0. 0. ]\n",
+ " [0. 0.00047662 0.03238289 ... 0.02048034 0. 0. ]\n",
+ " [0. 0.0005064 0.03374067 ... 0.01963579 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03971487 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00044683 0.03971487 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00044683 0.03971487 ... 0.01868567 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.03798371 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00049151 0.03910387 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03900204 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03482688 ... 0.01805226 0. 0. ]\n",
+ " [0. 0.00044683 0.03503055 ... 0.01781473 0. 0. ]\n",
+ " [0. 0.00044683 0.03503055 ... 0.01781473 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03659199 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.00047662 0.03170401 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.0311609 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03611677 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00047662 0.04025798 ... 0.01942465 0. 0. ]\n",
+ " [0. 0.00047662 0.0407332 ... 0.01931908 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.0305499 ... 0.01884402 0. 0. ]\n",
+ " [0. 0.00056598 0.03095723 ... 0.01894959 0. 0. ]\n",
+ " [0. 0.00047662 0.03109301 ... 0.01921351 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.02715547 ... 0.01894959 0. 0. ]\n",
+ " [0. 0.00047662 0.02715547 ... 0.01894959 0. 0. ]\n",
+ " [0. 0.00047662 0.02701969 ... 0.01894959 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03238289 ... 0.0192399 0. 0. ]\n",
+ " [0. 0.00049151 0.03360489 ... 0.01939826 0. 0. ]\n",
+ " [0. 0.00044683 0.03767821 ... 0.01908155 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00049151 0.03075356 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00049151 0.03075356 ... 0.0189232 0. 0. ]\n",
+ " [0. 0.00044683 0.02841141 ... 0.01908155 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.03957909 ... 0.01916073 0. 0. ]\n",
+ " [0. 0.0054513 0.03543788 ... 0.01942465 0. 0. ]\n",
+ " [0. 0.0005064 0.03665988 ... 0.01963579 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.04202308 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.04202308 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.04181942 ... 0.01900238 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.0734956053950727, 0.072789049140701, 0.07756887207465289, 0.08128593112375371, 0.07510316908234548, 0.09373327181570092, 0.06794275502103546, 0.07268098949301066, 0.05208059190583374, 0.049289015961857044] mean of list_RMSE_SousModele 0.07159692510139636\n",
+ " RMSE resultat vote 0.051218213871734135\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "key 1 value[6] 0.054938194722757405\n",
+ "key 8 value[6] 0.08787852233288299\n",
+ "key 11 value[6] 0.11586412658816056\n",
+ "key 12 value[6] 0.1288294285628787\n",
+ "key 13 value[6] 0.1444116730248319\n",
+ "key 14 value[6] 0.16334850821192484\n",
+ "key 17 value[6] 0.25465259711919624\n",
+ "key 18 value[6] 0.3081780004228992\n",
+ "key 19 value[6] 0.39107838594488675\n",
+ "key 20 value[6] 0.5474158342212556\n",
+ "key 21 value[6] 1.0\n",
+ "key 1 value[2] 0.0391616767585613\n",
+ "key 8 value[2] 0.0750675028170142\n",
+ "key 11 value[2] 0.11506590211207203\n",
+ "key 12 value[2] 0.16025205546557642\n",
+ "key 13 value[2] 0.21140179707975768\n",
+ "key 14 value[2] 0.2704633647101734\n",
+ "key 17 value[2] 0.3422195550582228\n",
+ "key 18 value[2] 0.4295985785411605\n",
+ "key 19 value[2] 0.5446954174267469\n",
+ "key 20 value[2] 0.707352415989607\n",
+ "key 21 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.057807377752721184\n",
+ "Current Error err_H 0.051218213871734135\n",
+ "myFeeder.t 21\n",
+ "[0.08582627654661987, 0.05673817506635284, 0.06456156236941377, 0.06281071100358247, 0.0888887469703822, 0.057653976383557515, 0.05978007937397294, 0.06264369230976617, 0.06201755783195607, 0.05860493048519744, 0.05515201309621722, 0.04648178257752944, 0.04672157710719657, 0.04980458195514284, 0.04642246206846767, 0.04025907547983262, 0.06302297235310445, 0.045869223165774825, 0.05168623143618591, 0.05779109135515785, 0.051218213871734135]\n"
+ ]
+ }
+ ],
+ "source": [
+ "\n",
+ "\n",
+ "path_X_Y_data = 'simulationDonnes/'\n",
+ "Y = np.load(path_X_Y_data + 'Y_CU_BEMS_19_normale_nettoy.npy' ,allow_pickle=True)\n",
+ "X = np.load(path_X_Y_data + 'X_CU_BEMS_19_normale_nettoybackTimePoint3PCA10.npy' ,allow_pickle=True)\n",
+ "\n",
+ " \n",
+ "while(True):\n",
+ " \n",
+ " dic_expert = {} #Conserver un dictionnaire des sous-modèles\n",
+ " timeList, errList, innErrList, testErrNextBatchList,nFlodErr= [],[],[],[],[] #Liste des temps, liste de la précision \n",
+ " Resulta_list, Ture_list = [],[] #Recorded information: predicted results and true values\n",
+ " dic_expert = {} #Conserver un dictionnaire des sous-modèles\n",
+ " ini_pourCentage = 0.01 # Le rapport initial doit être défini \n",
+ " testStar,testEnd = 0.2,0.99\n",
+ " HowManyDaysForBatch =14\n",
+ " batch_size = 1400*HowManyDaysForBatch# Taille du bloc\n",
+ " a,b = 0.35,1\n",
+ " time_Ini = 0 # Le temps nécessaire à l'initialisation est enregistré ici\n",
+ " varepsilon_extreme_bord = 1.2 # Coefficient permettant de vérifier si le sous-modèle fonctionne correctement\n",
+ " maxNumSousModle = 10 # Nombre maximal de sous-modèles\n",
+ " testBatchNum = 58 #Contrôler la quantité de circulation pendant l'expérience\n",
+ " Q3 = 75 # Définir les coefficients pour le calcul des poids d'instance\n",
+ " optionPCA, PCA_size = False, 25\n",
+ "# optitionInnErrOrCroissErr = 'InnErr'\n",
+ " optitionInnErrOrCroissErr = 'CroissErr'\n",
+ " optitionAddSousModele = True\n",
+ " optionVote ='Mean'\n",
+ " # optionVote ='OnlyMaxERRWeight'\n",
+ " # optionVote = 'vote_OnlyMaxESpWeight'\n",
+ " indicateur_sousModel = 'Random_LI_R_T_KNN' #['Random','LSTM','Lasso','KNN','R_Forest'] # 0 随机，1 LSTM，\n",
+ " cStepAugment = 1.4\n",
+ " updateOrNon = False\n",
+ " \n",
+ " myFeeder = feeder_Ini_Train_Batch(X,Y,testStar,testEnd,batch_size) # X,Y beginCentage,endCentage,batch_size \n",
+ " X_train,Y_train = myFeeder.getIni_X_Y(ini_pourCentage) #这里才设定初始比例\n",
+ " X_test,Y_test = myFeeder.getTrain_X_Y()\n",
+ "\n",
+ " iniFlage = True\n",
+ " while ((testBatchNum > myFeeder.t) and myFeeder.hasThisBatch()):\n",
+ "\n",
+ " # Créer un sous-modèle initialisé, code 0\n",
+ " start_Ini = time.process_time()# Heure de début d'enregistrement记录开始时间\n",
+ " if iniFlage == True:\n",
+ " Delta = [1/ myFeeder.batch_size]* myFeeder.batch_size #Valeur par défaut de delta delta的默认值 \n",
+ " dic_expert[0], err_numFlod= RandonSelectionModle(0,X_train, Y_train,optitionInnErrOrCroissErr, indique_sousModle = indicateur_sousModel)\n",
+ " iniFlage = False\n",
+ " time_Ini = round(time.process_time()-start_Ini,5) # Le temps nécessaire à l'initialisation est enregistré ici 这里记录了初始化需要的时间\n",
+ " start = time.process_time()# Heure de début d'enregistrement 记录开始时\n",
+ " \n",
+ " #----------------VOTE------------------------------------------------------\n",
+ " actul_Batch_X, actul_Batch_Y = myFeeder.getThisBatch()\n",
+ " # Calculer la performance de Ht-1 à t \n",
+ " if optionVote == 'Mean':\n",
+ " yhat_H = vote_mean(dic_expert,actul_Batch_X, actul_Batch_Y,afficher_detail=True)\n",
+ " elif optionVote == 'Max':\n",
+ " yhat_H = vote_OnlyMaxWeight(dic_expert,actul_Batch_X, actul_Batch_Y)\n",
+ "\n",
+ " print('vote is over')\n",
+ " #----------------------------------------------------------------------\n",
+ "\n",
+ " #----------------Confirm instance weights and record current cluster performance------------------------------------------------------\n",
+ " ERR_absolu_H = np.abs(actul_Batch_Y - yhat_H) #ERR_absolu (martix)\n",
+ " Resulta_list.append(yhat_H)\n",
+ " Ture_list.append(actul_Batch_Y)\n",
+ " err_H = aRMSE(actul_Batch_Y,yhat_H)# err as RMSE \n",
+ " meanERR_absolu_H = np.mean(ERR_absolu_H, axis=1)\n",
+ " # Delta = getMeanSuperPourCentage_Martix_ParLine(ERR_absolu_H,Q3)\n",
+ " # varepsilon_H = np.average(meanERR_absolu_H,weights=Delta)\n",
+ "\n",
+ " varepsilon_H = np.average(meanERR_absolu_H)\n",
+ " errList.append(err_H) \n",
+ " print('Instance weights, recording current group performance, over')\n",
+ " #----------------确认实例权重，记录当前群性能------------------------------------------------------\n",
+ "\n",
+ " #----------------Alarm in case of abnormality------------------------------------------------------\n",
+ " if err_H >10:\n",
+ " print('time: ',myFeeder.t,'Une erreur majeure s est produite：',err_H) \n",
+ " #----------------当群性能异常时，报警------------------------------------------------------\n",
+ "\n",
+ "\n",
+ " #------------------Add a new sub-model and confirm its weights (from cross-validation)-----------------------------\n",
+ " # 添加一个新的子模型 \n",
+ " # 这里有新改动，未在 meanERR_absolu < 0.1\n",
+ " bord = 1 \n",
+ " while(optitionAddSousModele):\n",
+ " print('Start adding new sub-models')\n",
+ " dic_expert[myFeeder.t], yhat_new_numFlod = RandonSelectionModle(myFeeder.t, actul_Batch_X, actul_Batch_Y, optitionInnErrOrCroissErr,indique_sousModle = indicateur_sousModel)\n",
+ "\n",
+ " \n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " ERR_absolu_new_numFlod = np.abs(actul_Batch_Y - yhat_new_numFlod) \n",
+ " meanERR_absolu_new_numFlod = np.mean(ERR_absolu_new_numFlod, axis=1)\n",
+ " err_new_numFlod = aRMSE(actul_Batch_Y,yhat_new_numFlod) \n",
+ " varepsilon_new = np.average(meanERR_absolu_new_numFlod,weights=Delta)\n",
+ " if varepsilon_new < varepsilon_H*bord:\n",
+ " nFlodErr.append(err_new_numFlod)\n",
+ " break\n",
+ " else:\n",
+ " bord = bord*cStepAugment\n",
+ " \n",
+ " elif optitionInnErrOrCroissErr == 'InnErr':\n",
+ "\n",
+ " yhat_new_inner = makePredictionModele(dic_expert[myFeeder.t][0], actul_Batch_X)\n",
+ " ERR_absolu_new_inner = np.abs(actul_Batch_Y - yhat_new_inner) \n",
+ " meanERR_absolu_new_inner = np.mean(ERR_absolu_new_inner, axis=1)\n",
+ " err_new_inner = aRMSE(actul_Batch_Y,yhat_new_inner) \n",
+ " varepsilon_new = np.average(meanERR_absolu_new_inner,weights=Delta)\n",
+ " break\n",
+ " #------------------添加一个新的子模型 ，确认其权重（来自交叉验证）-----------------------------\n",
+ " \n",
+ " if optitionAddSousModele:\n",
+ " #------------------Recording training errors for new sub-models-----------------------------\n",
+ " yhat_new_inner = makePredictionModele(dic_expert[myFeeder.t][0], actul_Batch_X)\n",
+ " err_new_inner = aRMSE(actul_Batch_Y,yhat_new_inner) \n",
+ " innErrList.append(err_new_inner)\n",
+ " #------------------对新的子模型记录训练错误-----------------------------\n",
+ "\n",
+ " #------------------Test error of the model generated by the previous block in the next block-----------------------------\n",
+ " if myFeeder.hasThisBatch_and_nextBath():\n",
+ " next_batch_test_X, next_batch_test_Y = myFeeder.getNextBatch_getThisBatch()\n",
+ " yhat_next_batch_test = makePredictionModele(dic_expert[myFeeder.t][0], next_batch_test_X)\n",
+ " err_test_NextBatch = aRMSE(next_batch_test_Y, yhat_next_batch_test) \n",
+ " testErrNextBatchList.append(err_test_NextBatch)\n",
+ " #------------------对上一个块的子模型在下一块的测试错误-----------------------------\n",
+ "\n",
+ "\n",
+ " #------------------Calculate the performance of the old sub-model based on the weighted strengths in the current block-----------\n",
+ " Varepsilon_list = [] #Collecte de la liste des pesées séparées de Varepsilon pour faciliter la recherche des valeurs max-min. 收集Varepsilon单独称list，方便找最大最小值\n",
+ " for key,value in dic_expert.items():\n",
+ " yhat_oldModel = makePredictionModele(value[0], actul_Batch_X)\n",
+ " ERR_absolu_oldModel = np.abs(actul_Batch_Y - yhat_oldModel)\n",
+ " mean_ERR_absolu_oldModel = np.mean(ERR_absolu_oldModel, axis=1)\n",
+ " err_oldModel = aRMSE(actul_Batch_Y,yhat_oldModel) # RMSE\n",
+ " varepsilon_oldModel = np.average(mean_ERR_absolu_oldModel,weights=Delta)\n",
+ " # if varepsilon_oldModel > varepsilon_H*varepsilon_extreme_bord: \n",
+ " # value[5] = varepsilon_H*varepsilon_extreme_bord\n",
+ " # else:\n",
+ " # value[5] = varepsilon_oldModel\n",
+ " value[5] = varepsilon_oldModel\n",
+ " Varepsilon_list.append(value[5])\n",
+ " value[7] = err_oldModel # enregrister RMSE\n",
+ " # for key,value in dic_expert.items(): #normalisation \n",
+ " # value[5] = (value[5] - min(Varepsilon_list))/(max(Varepsilon_list) - min(Varepsilon_list))\n",
+ " # for key,value in dic_expert.items(): \n",
+ " # print('key ', key, ' value[5] ', value[5] )\n",
+ " #------------------根据当前块中被加权的实力，计算旧的子模型的表现（7号普通错误），并更新权重（5号根据加权错误）-----------\n",
+ "\n",
+ "\n",
+ " #------------------Calculating time weights ----------\n",
+ " for key,value in dic_expert.items(): \n",
+ " numerator = beta_fonction(value[1],myFeeder.t,a,b)\n",
+ " denominator = 0\n",
+ " for j in range(0, myFeeder.t - value[1] +1): # \n",
+ " denominator += beta_fonction(myFeeder.t - j,myFeeder.t,a,b)\n",
+ "# print('denominator ' , denominator)\n",
+ "# if denominator == 0:\n",
+ "# print(' myFeeder.t ', myFeeder.t,' value[1]', value[1] )\n",
+ "# for j in range(0, myFeeder.t - value[1] +1): # \n",
+ "# print('j :', j , ' beta_fonction(myFeeder.t - j,myFeeder.t,a,b)', beta_fonction(myFeeder.t - j,myFeeder.t,a,b))\n",
+ " value[6] = numerator/denominator\n",
+ " for key,value in dic_expert.items(): \n",
+ " print('key ', key, ' value[6] ', value[6] )\n",
+ " #------------------计算6 Omega, 也就是时间权重 omega越小，越重要-----------\n",
+ "\n",
+ "\n",
+ " #------------------Calculating sub-model weights-----------\n",
+ " # Calculer les poids des sous-modèles\n",
+ " weight_list = []\n",
+ " for key,value in dic_expert.items(): \n",
+ " denominator = 0 \n",
+ " for j in range(0, myFeeder.t - value[1] +1): # from 0 to t-k\n",
+ " if ((myFeeder.t-j) in dic_expert.keys()):\n",
+ " denominator += dic_expert[myFeeder.t-j][6]*dic_expert[myFeeder.t-j][5]+0.1 # Omega 权重 * 5 当前块被加权的错误 \n",
+ " value[2] = math.log(1/denominator) \n",
+ " weight_list.append(math.log(1/denominator)) \n",
+ " for key,value in dic_expert.items(): \n",
+ " value[2] = (value[2] - min(weight_list)+0.1)/(max(weight_list) - min(weight_list)+0.1)\n",
+ " for key,value in dic_expert.items(): \n",
+ " print('key ', key, 'value[2] ',value[2] )\n",
+ " #------------------Calculating sub-model weights-----------\n",
+ "\n",
+ "\n",
+ " #------------------Removing the worst model -----------\n",
+ " if len(dic_expert) > maxNumSousModle:\n",
+ " maxErr = 0\n",
+ " del_key = 0\n",
+ " for key,value in dic_expert.items(): \n",
+ " if dic_expert[key][7] > maxErr:\n",
+ " maxErr = dic_expert[key][7]\n",
+ " del_key = key\n",
+ " dic_expert.pop(del_key)\n",
+ " #------------------去掉模型 最差策略 Sous-modèle de suppression de la pire stratégie-----------\n",
+ " \n",
+ " \n",
+ " #------------------Whether the sub-model LSTM is updated-----------\n",
+ " if updateOrNon:\n",
+ " updatingALLSousModele(dic_expert,actul_Batch_X,actul_Batch_Y) \n",
+ " #------------------子模型LSTM是否更新-----------\n",
+ "\n",
+ "\n",
+ " # 记录时间 Durée d'enregistrement\n",
+ " timeList.append(round(time.process_time()-start,3)) #记录时间\n",
+ "\n",
+ " print('Average errors so far np.mean(errList)', np.mean(errList)) #展示平均\n",
+ " print('Current Error err_H', err_H)\n",
+ "\n",
+ " #进入下一块\n",
+ " print('myFeeder.t ', myFeeder.t)\n",
+ " myFeeder.goNext()\n",
+ "\n",
+ " print(errList)\n",
+ "\n",
+ " break;\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.7.6"
+ },
+ "varInspector": {
+ "cols": {
+ "lenName": 16,
+ "lenType": 16,
+ "lenVar": 40
+ },
+ "kernels_config": {
+ "python": {
+ "delete_cmd_postfix": "",
+ "delete_cmd_prefix": "del ",
+ "library": "var_list.py",
+ "varRefreshCmd": "print(var_dic_list())"
+ },
+ "r": {
+ "delete_cmd_postfix": ") ",
+ "delete_cmd_prefix": "rm(",
+ "library": "var_list.r",
+ "varRefreshCmd": "cat(var_dic_list()) "
+ }
+ },
+ "position": {
+ "height": "662px",
+ "left": "601px",
+ "right": "20px",
+ "top": "121px",
+ "width": "800px"
+ },
+ "types_to_exclude": [
+ "module",
+ "function",
+ "builtin_function_or_method",
+ "instance",
+ "_Feature"
+ ],
+ "window_display": false
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/MORSTS.ipynb b/MORSTS.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..36bd7d3264269912123d346a1d1c2c679e690a3b
--- /dev/null
+++ b/MORSTS.ipynb
@@ -0,0 +1,3310 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import numpy as np\n",
+ "from sklearn.linear_model import LinearRegression\n",
+ "from sklearn.neighbors import KNeighborsRegressor\n",
+ "from sklearn.ensemble import RandomForestRegressor\n",
+ "from sklearn.tree import DecisionTreeRegressor\n",
+ "import os\n",
+ "from sklearn.metrics import mean_squared_error\n",
+ "import random\n",
+ "import math\n",
+ "import time\n",
+ "# from keras.models import Sequential\n",
+ "# from keras.layers import Dense\n",
+ "# from keras.layers import LSTM\n",
+ "from sklearn.decomposition import PCA\n",
+ "from sklearn.model_selection import KFold\n",
+ "import matplotlib.pyplot as plt\n",
+ "from random import choice\n",
+ "import pandas as pd\n",
+ "from sklearn.linear_model import Lasso\n",
+ "# from msvr import kernelmatrix\n",
+ "# from msvr import msvr"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# class MSVR_F:\n",
+ "# def __init__(self,ker = 'rbf', C = 2,epsi = 0.001,par = 0.8,tol = 1e-10):\n",
+ "# self.ker = ker\n",
+ "# self.C = C\n",
+ "# self.epsi = epsi\n",
+ "# self.par = par\n",
+ "# self.tol = tol\n",
+ " \n",
+ "# def fit(self,X_train,Y_train):\n",
+ "# X_train = np.array(X_train)\n",
+ "# Y_train = np.array(Y_train)\n",
+ "# self.X_train = X_train\n",
+ "# self.Beta = msvr(X_train, Y_train, self.ker, self.C, self.epsi, self.par, self.tol)\n",
+ " \n",
+ "# def predict(self,X_test):\n",
+ "# X_test = np.array(X_test)\n",
+ "# H = kernelmatrix('rbf', X_test, self.X_train, par);\n",
+ "# return np.dot(H, self.Beta)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Ecrire un générateur de blocs pour obtenir différents ensembles de données initiales\n",
+ "# data est la donnée, n est le nombre de parties en lesquelles elle est divisée, i est la première\n",
+ "# 写一个块生成器，用来获得不同的初始数据集\n",
+ "# data 是数据，n是分成几份，i是第几份\n",
+ "\n",
+ "def getChunks(data,n,i):\n",
+ " return data[ int(len(data)*(i)/n)+1 : (int(len(data)*(i+1)/n)-1)]\n",
+ "def getPourCentage(data,beginCentage,endCentage):\n",
+ " return data[ (int(len(data)*beginCentage)): (int(len(data)*endCentage))]\n",
+ "#功能介绍：返回初始数据，返回全部训练数据，判断是否还有下一个batch，返回当前batch，\n",
+ "#Description de la fonction : retour des données initiales, retour de toutes les données d'entraînement, détermination de l'existence d'un lot suivant, retour du lot actuel.\n",
+ "class feeder_Ini_Train_Batch():\n",
+ " def __init__(self,X,Y,beginCentage,endCentage,batch_size):\n",
+ " self.X = X\n",
+ " self.Y = Y\n",
+ " #elf.X_train = getPourCentage(X,0,iniCentage)\n",
+ " #elf.Y_train = getPourCentage(Y,0,iniCentage)\n",
+ " self.X_test = getPourCentage(X,beginCentage,endCentage)\n",
+ " self.Y_test = getPourCentage(Y,beginCentage,endCentage)\n",
+ " self.batch_size = batch_size\n",
+ " self.t = 1 # t C'est l'indexation et le temps. 就是索引和时间\n",
+ " def getIni_X_Y(self,iniCentage):\n",
+ " return getPourCentage(self.X,0,iniCentage),getPourCentage(self.Y,0,iniCentage)\n",
+ " def getTrain_X_Y(self):\n",
+ " return self.X_test,self.Y_test\n",
+ " def hasThisBatch(self):\n",
+ " if (self.t)*self.batch_size < len (self.X_test):\n",
+ " return True\n",
+ " else:\n",
+ " return False\n",
+ " def hasThisBatch_and_nextBath(self):\n",
+ " if (self.t+1)*(self.batch_size) < len (self.X_test):\n",
+ " return True\n",
+ " else:\n",
+ " return False\n",
+ " def getThisBatch(self):\n",
+ " if self.hasThisBatch() == True:\n",
+ " actul_Batch_X = self.X_test[(self.t-1)*self.batch_size:(self.t)*self.batch_size]\n",
+ " actul_Batch_Y = self.Y_test[(self.t-1)*self.batch_size:(self.t)*self.batch_size]\n",
+ " return actul_Batch_X,actul_Batch_Y\n",
+ " else:\n",
+ " print('err index out')\n",
+ " def getNextBatch_getThisBatch(self):\n",
+ " if self.hasThisBatch() == True:\n",
+ " next_Batch_X = self.X_test[(self.t)*self.batch_size:(self.t+1)*self.batch_size]\n",
+ " next_Batch_Y = self.Y_test[(self.t)*self.batch_size:(self.t+1)*self.batch_size]\n",
+ " return next_Batch_X,next_Batch_Y\n",
+ " else:\n",
+ " print('err index out')\n",
+ " def goNext(self):\n",
+ " self.t +=1\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def getMeanSuperPourCentage(data,PourCentage): # Calculez la valeur moyenne du pourcentage supérieur des données. 计算排名前百分之多少的数据的均值\n",
+ "# print(np.percentile(data, (PourCentage)))\n",
+ " return np.mean([x for x in data if x >= np.percentile(data, (PourCentage))])\n",
+ "def getMeanSuperPourCentage_Martix_ParLine(data,PourCentage):\n",
+ " return np.array([getMeanSuperPourCentage(x,PourCentage) for x in data ])\n",
+ "\n",
+ " # 0子模型 1编号时间 2权重 3class名 4俗称 5varepsilon加权错 6 Omega时间权重 7err普通RMSE错误 8 模型地址\n",
+ " # 0 submodèle 1 temps de numérotation 2 poids 3 nom de classe 4 nom commun 5 erreur de pondération varepsilon 6 pondération temps oméga 7 erreur RMSE commune 8 adresse du modèle\n",
+ "\n",
+ "def RandonSelectionModle(index,x,y , optitionInnErrOrCroissErr,numFlod = 3, indique_sousModle = 'Random_LS_LI_R_T'): \n",
+ "\n",
+ " if indique_sousModle == 'Random_LI_R_T_KNN':\n",
+ " indique_sousModle = choice(['Lasso','R_Forest','Tree','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_R_T_KNN':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','R_Forest','Tree','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_R_T_SVR':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','R_Forest','Tree','SVR'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_R_T':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','R_Forest','Tree'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_R_KNN':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','R_Forest','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_T_KNN':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','Tree','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_R_T_KNN':\n",
+ "# indique_sousModle = choice(['LSTM','R_Forest','Tree','KNN'])\n",
+ "# elif indique_sousModle == 'Random_LS_R_T':\n",
+ "# indique_sousModle = choice(['LSTM','R_Forest','Tree'])\n",
+ "# elif indique_sousModle == 'Random_LS_LI_T':\n",
+ "# indique_sousModle = choice(['LSTM','Lasso','Tree'])\n",
+ " \n",
+ " print('Randomly selected indique_sousModle ',indique_sousModle)\n",
+ " weight = 1 # weight of this base model 投票权重\n",
+ " varepsilon = 1 # Résultats des fonctions de perte 丢失函数结果 也就是 varepsilon 加权之后的错\n",
+ " omega = 1 # la pondération du temps 关于时间的权重\n",
+ " RMSE = 1 # err 普通RMSE错误\n",
+ " x_rnn = np.reshape(x, (x.shape[0], 1,x.shape[1]))\n",
+ " kf = KFold(n_splits=numFlod)\n",
+ " yhat_all_flod = []\n",
+ " \n",
+ "\n",
+ " \n",
+ "# elif indique_sousModle == 'LSTM' :\n",
+ "# modele = Sequential()\n",
+ "# modele.add(LSTM(500, input_shape =(1, x_rnn.shape[2]) , activation='relu'))\n",
+ "# modele.add(Dense(500, activation='relu'))\n",
+ "# modele.add(Dense(y.shape[1] , activation='sigmoid'))\n",
+ "# # Compile model\n",
+ "# modele.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
+ "# modele.fit(x_rnn, y, epochs=6, batch_size=2000, verbose=2)\n",
+ "# if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ "# for train, test in kf.split(x_rnn):\n",
+ "# X_train_numFlod = np.array( [x_rnn[i] for i in train])\n",
+ "# Y_train_numFlod = np.array([y[i] for i in train])\n",
+ "# X_test_numFlod = np.array([x_rnn[i] for i in test])\n",
+ "# Y_test_numFlod = np.array([y[i] for i in test])\n",
+ "# modele_numFlod = Sequential()\n",
+ "# modele_numFlod.add(LSTM(500, input_shape =(1, x_rnn.shape[2]) , activation='relu'))\n",
+ "# modele_numFlod.add(Dense(500, activation='relu'))\n",
+ "# modele_numFlod.add(Dense(y.shape[1] , activation='sigmoid'))\n",
+ "# modele_numFlod.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
+ "# modele_numFlod.fit(X_train_numFlod, Y_train_numFlod, epochs=4, batch_size=1000, verbose=2)\n",
+ "# yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ "# yhat_all_flod += yhat__numFlod.tolist()\n",
+ "# return [modele,index,weight,modele.__class__,'LSTM',varepsilon,omega,RMSE,id(modele)] , yhat_all_flod\n",
+ " \n",
+ " \n",
+ " \n",
+ "# elif indique_sousModle == 'SVR' :\n",
+ "# modele = MSVR_F()\n",
+ "# modele.fit(x,y)\n",
+ "# if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ "# for train, test in kf.split(x):\n",
+ "# X_train_numFlod = np.array( [x[i] for i in train])\n",
+ "# Y_train_numFlod = np.array([y[i] for i in train])\n",
+ "# X_test_numFlod = np.array([x[i] for i in test])\n",
+ "# Y_test_numFlod = np.array([y[i] for i in test])\n",
+ "# modele_numFlod = MSVR_F()\n",
+ "# modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ "# yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ "# yhat_all_flod += yhat__numFlod.tolist() \n",
+ "# return [modele,index,weight,modele.__class__,'SVR',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ "\n",
+ " if indique_sousModle == 'KNN':\n",
+ " modele = KNeighborsRegressor(n_neighbors=15)\n",
+ " modele.fit(x,y)\n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " for train, test in kf.split(x):\n",
+ " X_train_numFlod = np.array( [x[i] for i in train])\n",
+ " Y_train_numFlod = np.array([y[i] for i in train])\n",
+ " X_test_numFlod = np.array([x[i] for i in test])\n",
+ " Y_test_numFlod = np.array([y[i] for i in test])\n",
+ " modele_numFlod = KNeighborsRegressor(n_neighbors=15)\n",
+ " modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ " yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ " yhat_all_flod += yhat__numFlod.tolist() \n",
+ " return [modele,index,weight,modele.__class__,'KNeighborsRegressor',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ " elif indique_sousModle == 'R_Forest':\n",
+ " modele = RandomForestRegressor(n_estimators=10, )\n",
+ " modele.fit(x,y)\n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " for train, test in kf.split(x):\n",
+ " X_train_numFlod = np.array( [x[i] for i in train])\n",
+ " Y_train_numFlod = np.array([y[i] for i in train])\n",
+ " X_test_numFlod = np.array([x[i] for i in test])\n",
+ " Y_test_numFlod = np.array([y[i] for i in test])\n",
+ " modele_numFlod = RandomForestRegressor(n_estimators=10,)\n",
+ " modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ " yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ " yhat_all_flod += yhat__numFlod.tolist()\n",
+ " return [modele,index,weight,modele.__class__,'RandomForestRegressor',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ " elif indique_sousModle == 'Lasso':\n",
+ " modele = Lasso(alpha=0.1)\n",
+ " modele.fit(x,y)\n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " for train, test in kf.split(x):\n",
+ " X_train_numFlod = np.array( [x[i] for i in train])\n",
+ " Y_train_numFlod = np.array([y[i] for i in train])\n",
+ " X_test_numFlod = np.array([x[i] for i in test])\n",
+ " Y_test_numFlod = np.array([y[i] for i in test])\n",
+ " modele_numFlod = Lasso(alpha=0.1)\n",
+ " modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ " yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ " yhat_all_flod += yhat__numFlod.tolist() \n",
+ " return [modele,index,weight,modele.__class__,'LassoRegression',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ " elif indique_sousModle == 'Tree':\n",
+ " modele = DecisionTreeRegressor()\n",
+ " modele.fit(x,y)\n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " for train, test in kf.split(x):\n",
+ " X_train_numFlod = np.array( [x[i] for i in train])\n",
+ " Y_train_numFlod = np.array([y[i] for i in train])\n",
+ " X_test_numFlod = np.array([x[i] for i in test])\n",
+ " Y_test_numFlod = np.array([y[i] for i in test])\n",
+ " modele_numFlod = DecisionTreeRegressor()\n",
+ " modele_numFlod.fit(X_train_numFlod,Y_train_numFlod)\n",
+ " yhat__numFlod = modele_numFlod.predict(X_test_numFlod)\n",
+ " yhat_all_flod += yhat__numFlod.tolist() \n",
+ " return [modele,index,weight,modele.__class__,'DecisionTreeRegressor',varepsilon,omega,RMSE,id(modele)],yhat_all_flod\n",
+ " \n",
+ "def makePredictionModele(modele,x):\n",
+ " if type(modele) in [type(RandomForestRegressor()), \n",
+ " type(KNeighborsRegressor()),\n",
+ " type(Lasso()),\n",
+ " type(DecisionTreeRegressor())] :\n",
+ "# type(MSVR_F())\n",
+ "\n",
+ " return modele.predict(x)\n",
+ " else:\n",
+ " print('报错，不在列表中 Error reported, not in the list')\n",
+ "\n",
+ "def beta_fonction(k,t,a,b): \n",
+ " return 1/(1+ math.exp(-a*(t-k-b))) \n",
+ "\n",
+ "def vote_mean(dic_expert, actul_Batch_X, actul_Batch_Y, afficher_detail = False):\n",
+ " H_res = np.array( [[0.0]* len(actul_Batch_Y[0])] *len(actul_Batch_Y))\n",
+ " print(H_res)\n",
+ " sumWeigt = 0\n",
+ " \n",
+ " list_RMSE_SousModele = []\n",
+ "\n",
+ " for key,value in dic_expert.items():\n",
+ " yhat_sousM = makePredictionModele(value[0],actul_Batch_X)\n",
+ " \n",
+ " if np.array(yhat_sousM).ndim == 1:\n",
+ " print('type([[z] for z in yhat_sousM]) ',type([[z] for z in yhat_sousM]))\n",
+ " yhat_sousModel = np.array([[z] for z in yhat_sousM])\n",
+ " else:\n",
+ " yhat_sousModel = yhat_sousM\n",
+ " print(type(yhat_sousModel))\n",
+ " print('Output of each sub-model',yhat_sousModel)\n",
+ " H_res =H_res + yhat_sousModel*value[2]\n",
+ " sumWeigt += value[2] # 2 représente les poids des votes des sous-modèles 2号代表子模型投票权\n",
+ " list_RMSE_SousModele.append(aRMSE( actul_Batch_Y ,yhat_sousModel))\n",
+ " yhat_H = np.array(H_res)/sumWeigt # L'obtention des résultats prédits pour H\n",
+ " \n",
+ "# print('投票结束之后的输出', yhat_H)\n",
+ "# print('Y的原始值', actul_Batch_Y)\n",
+ " if afficher_detail == True:\n",
+ " print( 'list_RMSE_SousModele ', list_RMSE_SousModele, 'mean of list_RMSE_SousModele', np.mean(list_RMSE_SousModele) )\n",
+ " print(' RMSE resultat vote', aRMSE( actul_Batch_Y ,yhat_H))\n",
+ " return yhat_H\n",
+ "\n",
+ "\n",
+ "def vote_OnlyMaxERRWeight(dic_expert, actul_Batch_X, actul_Batch_Y):\n",
+ " the_key = 0\n",
+ " maxWeight = 0\n",
+ " for key,value in dic_expert.items():\n",
+ " if dic_expert[key][5] > maxWeight:\n",
+ " maxWeight = dic_expert[key][5]\n",
+ " the_key = key\n",
+ " yhat_H = makePredictionModele(dic_expert[the_key][0],actul_Batch_X)\n",
+ " err_H = aRMSE(actul_Batch_Y,yhat_H)# err as RMSE \n",
+ " print('Sub-model selected to vote, no. ', the_key, ' 其 RMSE', err_H)\n",
+ " return yhat_H\n",
+ "\n",
+ "def vote_OnlyMaxESpWeight(dic_expert, actul_Batch_X, actul_Batch_Y):\n",
+ " the_key = 0\n",
+ " maxWeight = 0\n",
+ " for key,value in dic_expert.items():\n",
+ " if dic_expert[key][2] > maxWeight:\n",
+ " maxWeight = dic_expert[key][2]\n",
+ " the_key = key\n",
+ " yhat_H = makePredictionModele(dic_expert[the_key][0],actul_Batch_X)\n",
+ " err_H = aRMSE(actul_Batch_Y,yhat_H)# err as RMSE \n",
+ " print('Sub-model selected to vote, no. ', the_key, ' with its RMSE', err_H)\n",
+ " return yhat_H\n",
+ "\n",
+ "def updatingALLSousModele(dic_expert, x, y):\n",
+ " print (' Update sub-model program start' )\n",
+ " x_rnn = np.reshape(x, (x.shape[0], 1,x.shape[1]))\n",
+ " for key,value in dic_expert.items():\n",
+ " if type(value[0]) in [type(Sequential())]:\n",
+ " value[0].fit(x_rnn, actul_Batch_Y)\n",
+ "\n",
+ " \n",
+ "def get_filename(path,filetype): # 输入路径、文件类型例如'.csv'\n",
+ " name = []\n",
+ " for root,dirs,files in os.walk(path):\n",
+ " for i in files:\n",
+ " if os.path.splitext(i)[1]==filetype:\n",
+ " name.append(i) \n",
+ " return name\n",
+ "\n",
+ "def aRMSE(y_true,y_pred):\n",
+ " return np.mean(mean_squared_error(y_true, y_pred, squared=False, multioutput='raw_values'))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Randomly selected indique_sousModle R_Forest\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00831099 0.01283096 ... 0.01765637 0. 0. ]\n",
+ " [0. 0.01094727 0.01120163 ... 0.01765637 0. 0. ]\n",
+ " [0.04600991 0.07971403 0.07494908 ... 0.0202692 0.02935117 0.10079576]\n",
+ " ...\n",
+ " [0.41287837 0.36769437 0.32830957 ... 0.03095804 0.1267486 0.44482759]\n",
+ " [0.38292057 0.36858803 0.3089613 ... 0.03024545 0.1236942 0.45092838]\n",
+ " [0.46006237 0.39687221 0.33991853 ... 0.02961203 0.11684439 0.42785146]]\n",
+ "list_RMSE_SousModele [0.09649610570283121] mean of list_RMSE_SousModele 0.09649610570283121\n",
+ " RMSE resultat vote 0.09649610570283121\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 0 value[6] 0.5474158342212556\n",
+ "key 1 value[6] 1.0\n",
+ "key 0 value[2] 0.11420994880784559\n",
+ "key 1 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.09649610570283121\n",
+ "Current Error err_H 0.09649610570283121\n",
+ "myFeeder.t 1\n",
+ "[0.09649610570283121]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.44311136 0.39562109 0.32892057 ... 0.02897862 0.12317007 0.41246684]\n",
+ " [0.50330215 0.37774799 0.33228106 ... 0.03151227 0.13392373 0.45331565]\n",
+ " [0.50330215 0.37774799 0.33228106 ... 0.03151227 0.13392373 0.45331565]\n",
+ " ...\n",
+ " [0. 0.00317248 0.01639511 ... 0.01828979 0. 0. ]\n",
+ " [0. 0.00317248 0.01639511 ... 0.01828979 0. 0. ]\n",
+ " [0. 0.00321716 0.01680244 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.79245398e-01 3.60321716e-01 3.36456212e-01 ... 3.10899974e-02\n",
+ " 1.61829026e-01 5.12643678e-01]\n",
+ " [5.86828105e-01 3.65534704e-01 3.40868975e-01 ... 3.53127474e-02\n",
+ " 1.63961684e-01 5.20601238e-01]\n",
+ " [5.87146089e-01 3.65921954e-01 3.41208418e-01 ... 3.53655318e-02\n",
+ " 1.64046027e-01 5.20954907e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 2.82416836e-02 ... 1.87384534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.68160217e-02 ... 1.86328847e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.25390360e-02 ... 1.85801003e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.10076198299505745, 0.05665463313893739] mean of list_RMSE_SousModele 0.07870830806699743\n",
+ " RMSE resultat vote 0.056939832739505786\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "key 0 value[6] 0.39107838594488675\n",
+ "key 1 value[6] 0.5474158342212556\n",
+ "key 2 value[6] 1.0\n",
+ "key 0 value[2] 0.07603479250022051\n",
+ "key 1 value[2] 0.42447746523848895\n",
+ "key 2 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.0767179692211685\n",
+ "Current Error err_H 0.056939832739505786\n",
+ "myFeeder.t 2\n",
+ "[0.09649610570283121, 0.056939832739505786]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00321716 0.01680244 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00576408 0.01486762 ... 0.01828979 0. 0. ]\n",
+ " [0. 0.00446828 0.01588595 ... 0.01828979 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.01613047 0.01456212 ... 0.01955661 0. 0. ]\n",
+ " [0. 0.00335121 0.01731161 ... 0.01979414 0. 0. ]\n",
+ " [0. 0.00335121 0.01761711 ... 0.01963579 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02111337 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00047662 0.0245757 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00053619 0.02939579 ... 0.01879124 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.01900883 ... 0.02179995 0. 0. ]\n",
+ " [0. 0.00047662 0.01832994 ... 0.02211665 0. 0. ]\n",
+ " [0. 0.00044683 0.01900883 ... 0.02185273 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02545825 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02545825 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02647658 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02545825 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01832994 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.02545825 ... 0.01821061 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.10388930906488841, 0.06015379098027394, 0.09576844315376118] mean of list_RMSE_SousModele 0.08660384773297451\n",
+ " RMSE resultat vote 0.07351295895389018\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 0 value[6] 0.3081780004228992\n",
+ "key 1 value[6] 0.39107838594488675\n",
+ "key 2 value[6] 0.5474158342212556\n",
+ "key 3 value[6] 1.0\n",
+ "key 0 value[2] 0.06366816530241602\n",
+ "key 1 value[2] 0.2583600150703267\n",
+ "key 2 value[2] 0.5112479928123908\n",
+ "key 3 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07564963246540905\n",
+ "Current Error err_H 0.07351295895389018\n",
+ "myFeeder.t 3\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01470063 0.01782077 ... 0.01916073 0. 0. ]\n",
+ " [0. 0.02310098 0.02107943 ... 0.02011085 0. 0. ]\n",
+ " [0. 0.00196604 0.01863544 ... 0.02019002 0. 0. ]\n",
+ " ...\n",
+ " [0.48433315 0.31264522 0.26812627 ... 0.02953286 0.11886861 0.39071618]\n",
+ " [0.48433315 0.31264522 0.26812627 ... 0.02953286 0.11886861 0.39071618]\n",
+ " [0.48433315 0.31264522 0.26812627 ... 0.02953286 0.11886861 0.39071618]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.05023761e-02 ... 2.19055160e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.12491514e-02 ... 2.20638691e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.04344874e-02 ... 2.20110847e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.57701951e-01 3.28060769e-01 3.19076714e-01 ... 2.93481130e-02\n",
+ " 1.94047834e-01 4.41909814e-01]\n",
+ " [6.59940072e-01 3.28388442e-01 3.18397828e-01 ... 2.94008973e-02\n",
+ " 1.93734562e-01 4.44385500e-01]\n",
+ " [6.39907051e-01 3.25141495e-01 3.09572301e-01 ... 2.88730536e-02\n",
+ " 1.93734562e-01 4.40671972e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 1.34048257e-03 2.03665988e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.54582485e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.54582485e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [7.69216657e-01 3.23949955e-01 2.58655804e-01 ... 3.24623911e-02\n",
+ " 1.98264956e-01 4.90716180e-01]\n",
+ " [7.69216657e-01 3.23949955e-01 2.58655804e-01 ... 3.24623911e-02\n",
+ " 1.98264956e-01 4.90716180e-01]\n",
+ " [7.69216657e-01 3.23949955e-01 2.58655804e-01 ... 3.24623911e-02\n",
+ " 1.98264956e-01 4.90716180e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.71894094e-02 ... 2.12984956e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 2.58655804e-02 ... 1.96357878e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.63747454e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [7.17538066e-01 3.40661305e-01 3.29124236e-01 ... 2.99287411e-02\n",
+ " 1.99313212e-01 4.85941645e-01]\n",
+ " [7.45367822e-01 3.42180518e-01 3.12219959e-01 ... 3.11955661e-02\n",
+ " 1.99240918e-01 4.92307692e-01]\n",
+ " [7.53439736e-01 3.34897230e-01 3.08350305e-01 ... 3.03246239e-02\n",
+ " 2.01174770e-01 4.87798408e-01]]\n",
+ "list_RMSE_SousModele [0.12525790318528365, 0.07376921430269853, 0.10941083493015667, 0.06361311253681644] mean of list_RMSE_SousModele 0.09301276623873883\n",
+ " RMSE resultat vote 0.06551391490994131\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 0 value[6] 0.25465259711919624\n",
+ "key 1 value[6] 0.3081780004228992\n",
+ "key 2 value[6] 0.39107838594488675\n",
+ "key 3 value[6] 0.5474158342212556\n",
+ "key 4 value[6] 1.0\n",
+ "key 0 value[2] 0.0560442918595943\n",
+ "key 1 value[2] 0.18915569053102202\n",
+ "key 2 value[2] 0.34739356056306586\n",
+ "key 3 value[2] 0.5874034422897825\n",
+ "key 4 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07311570307654212\n",
+ "Current Error err_H 0.06551391490994131\n",
+ "myFeeder.t 4\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.48380114 0.30290438 0.26619145 ... 0.03016627 0.11865173 0.39973475]\n",
+ " [0.4916896 0.30799821 0.27403259 ... 0.03079968 0.10267486 0.40212202]\n",
+ " [0.48174647 0.30299374 0.26547862 ... 0.03016627 0.11586843 0.39973475]\n",
+ " ...\n",
+ " [0. 0.01219839 0.00977597 ... 0.01797308 0. 0. ]\n",
+ " [0. 0.00844504 0.01283096 ... 0.01836896 0. 0. ]\n",
+ " [0. 0.00969616 0.01221996 ... 0.01797308 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.16951018e-01 3.20226393e-01 2.99185336e-01 ... 2.78701504e-02\n",
+ " 1.93288752e-01 4.43678161e-01]\n",
+ " [6.16951018e-01 3.20226393e-01 2.99185336e-01 ... 2.78701504e-02\n",
+ " 1.93288752e-01 4.43678161e-01]\n",
+ " [6.18296337e-01 3.23056300e-01 2.99049559e-01 ... 2.78701504e-02\n",
+ " 1.92951383e-01 4.50751547e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 2.62050238e-02 ... 1.92135128e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.60013578e-02 ... 1.92135128e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.58655804e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.76921666 0.32394996 0.2586558 ... 0.03246239 0.19826496 0.49071618]\n",
+ " [0.76921666 0.32394996 0.2586558 ... 0.03246239 0.19826496 0.49071618]\n",
+ " [0.76921666 0.32394996 0.2586558 ... 0.03246239 0.19826496 0.49071618]\n",
+ " ...\n",
+ " [0. 0.00089366 0.03258656 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.00089366 0.03258656 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.00089366 0.03258656 ... 0.02137767 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.68060906e-01 3.44906166e-01 3.13136456e-01 ... 3.15914489e-02\n",
+ " 2.00198807e-01 4.91246684e-01]\n",
+ " [7.68574573e-01 3.43967828e-01 3.12118126e-01 ... 3.09580364e-02\n",
+ " 1.99656606e-01 4.91246684e-01]\n",
+ " [7.66813429e-01 3.46470063e-01 3.13441955e-01 ... 3.16706255e-02\n",
+ " 1.99512019e-01 4.96021220e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.91510277e-04 2.51527495e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.53564155e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 2.46435845e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.39919281e-01 3.36684540e-01 3.28309572e-01 ... 3.50752177e-02\n",
+ " 1.98228809e-01 4.36339523e-01]\n",
+ " [7.35589800e-01 3.26720286e-01 3.23116090e-01 ... 3.50752177e-02\n",
+ " 1.98716790e-01 4.27055703e-01]\n",
+ " [7.70757659e-01 3.25737265e-01 3.25967413e-01 ... 3.49168646e-02\n",
+ " 1.99060184e-01 4.25994695e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.91510277e-04 1.80244399e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 1.77189409e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 1.87372709e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.1085689748836216, 0.09137311316122942, 0.1253711821585215, 0.09994759191654855, 0.10725710614293185] mean of list_RMSE_SousModele 0.10650359365257059\n",
+ " RMSE resultat vote 0.09254022964979645\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 0 value[6] 0.21615527050978905\n",
+ "key 1 value[6] 0.25465259711919624\n",
+ "key 2 value[6] 0.3081780004228992\n",
+ "key 3 value[6] 0.39107838594488675\n",
+ "key 4 value[6] 0.5474158342212556\n",
+ "key 5 value[6] 1.0\n",
+ "key 0 value[2] 0.0531076763982407\n",
+ "key 1 value[2] 0.14649025812989674\n",
+ "key 2 value[2] 0.25612627275943944\n",
+ "key 3 value[2] 0.4029064792038676\n",
+ "key 4 value[2] 0.610983898993079\n",
+ "key 5 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07700060839119299\n",
+ "Current Error err_H 0.09254022964979645\n",
+ "myFeeder.t 5\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00965147 0.01221996 ... 0.01797308 0. 0. ]\n",
+ " [0. 0.01219839 0.0101833 ... 0.01733967 0. 0. ]\n",
+ " [0. 0.01103664 0.0107943 ... 0.01773555 0. 0. ]\n",
+ " ...\n",
+ " [0.46718033 0.37086685 0.33767821 ... 0.03127474 0.11713356 0.48196286]\n",
+ " [0.46718033 0.37086685 0.33767821 ... 0.03127474 0.11713356 0.48196286]\n",
+ " [0.35742066 0.36251117 0.2898167 ... 0.02794933 0.12163383 0.45013263]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.65981531e-04 2.28784793e-02 ... 1.87384534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.00950441e-02 ... 1.84745315e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.96877122e-02 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.90166942e-01 3.51921358e-01 3.28716904e-01 ... 2.77117973e-02\n",
+ " 1.89312609e-01 5.01503095e-01]\n",
+ " [5.84638904e-01 3.52874590e-01 3.38628649e-01 ... 2.77117973e-02\n",
+ " 1.88204109e-01 5.01326260e-01]\n",
+ " [6.27848101e-01 3.53500149e-01 3.31093007e-01 ... 2.97703880e-02\n",
+ " 1.82203747e-01 4.94960212e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00089366 0.03258656 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.00089366 0.03258656 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.00089366 0.03258656 ... 0.02137767 0. 0. ]\n",
+ " ...\n",
+ " [0.58099431 0.37399464 0.27698574 ... 0.03483769 0.19338514 0.54111406]\n",
+ " [0.78132453 0.36952636 0.35539715 ... 0.02692003 0.19555395 0.5066313 ]\n",
+ " [0.74224913 0.36327078 0.32688391 ... 0.03562945 0.18868607 0.55437666]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.36193029e-04 2.42362525e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.55600815e-02 ... 1.92399050e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.52545825e-02 ... 1.92399050e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.65987892e-01 3.63941019e-01 3.17617108e-01 ... 3.37292162e-02\n",
+ " 1.77299837e-01 5.04509284e-01]\n",
+ " [7.49660613e-01 3.36193029e-01 2.68940937e-01 ... 3.06413302e-02\n",
+ " 1.92698355e-01 4.72679045e-01]\n",
+ " [7.35443038e-01 3.54200179e-01 3.33197556e-01 ... 3.44418052e-02\n",
+ " 1.91053678e-01 5.36870027e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 1.73116090e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 1.61914460e-02 ... 1.93982581e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.80875782e-04 1.68024440e-02 ... 1.93982581e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.68666300e-01 3.55942806e-01 3.27494908e-01 ... 3.27790974e-02\n",
+ " 1.93873125e-01 4.71618037e-01]\n",
+ " [6.52339020e-01 3.46023235e-01 3.18839104e-01 ... 3.38875693e-02\n",
+ " 1.76613049e-01 4.27055703e-01]\n",
+ " [6.51953770e-01 3.56166220e-01 3.38391039e-01 ... 3.30958036e-02\n",
+ " 1.96945599e-01 4.15384615e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.06404528e-04 2.54582485e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.54582485e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.54582485e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [7.31670030e-01 3.35358951e-01 3.32315003e-01 ... 3.54183162e-02\n",
+ " 1.96614254e-01 4.41025641e-01]\n",
+ " [7.71601541e-01 3.56419422e-01 3.44399185e-01 ... 3.13539192e-02\n",
+ " 1.94505693e-01 4.35720601e-01]\n",
+ " [7.48535437e-01 3.42389038e-01 3.40325866e-01 ... 3.15122724e-02\n",
+ " 1.79312007e-01 4.29177719e-01]]\n",
+ "list_RMSE_SousModele [0.09619322972575758, 0.07362281360063773, 0.11241726876060101, 0.08305472665856346, 0.0899183111205803, 0.05384648929203804] mean of list_RMSE_SousModele 0.08484213985969635\n",
+ " RMSE resultat vote 0.05907000031746698\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "key 0 value[6] 0.18670485439918758\n",
+ "key 1 value[6] 0.21615527050978905\n",
+ "key 2 value[6] 0.25465259711919624\n",
+ "key 3 value[6] 0.3081780004228992\n",
+ "key 4 value[6] 0.39107838594488675\n",
+ "key 5 value[6] 0.5474158342212556\n",
+ "key 6 value[6] 1.0\n",
+ "key 0 value[2] 0.04659667188146264\n",
+ "key 1 value[2] 0.1190577879150322\n",
+ "key 2 value[2] 0.20169683802819963\n",
+ "key 3 value[2] 0.30770053980475376\n",
+ "key 4 value[2] 0.44485247395821864\n",
+ "key 5 value[2] 0.6496652851381268\n",
+ "key 6 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07401217371223866\n",
+ "Current Error err_H 0.05907000031746698\n",
+ "myFeeder.t 6\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.40955788 0.36318141 0.32148676 ... 0.03159145 0.11463944 0.45941645]\n",
+ " [0.42891213 0.36076854 0.32688391 ... 0.03159145 0.11447678 0.44748011]\n",
+ " [0.45648505 0.38941019 0.32912424 ... 0.02897862 0.14059281 0.43023873]\n",
+ " ...\n",
+ " [0. 0.0642538 0.06242363 ... 0.02779097 0. 0. ]\n",
+ " [0.1282517 0.12743521 0.11517312 ... 0.03087886 0.04534611 0.14827586]\n",
+ " [0.16272244 0.14307417 0.12311609 ... 0.02620744 0.03430327 0.15092838]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.58345258 0.35251713 0.34175153 ... 0.02713117 0.18803542 0.49496021]\n",
+ " [0.60374243 0.35117665 0.33102512 ... 0.02744788 0.18231219 0.49867374]\n",
+ " [0.58211949 0.35582365 0.34209097 ... 0.02845078 0.18132418 0.50044209]\n",
+ " ...\n",
+ " [0. 0.08087578 0.05485404 ... 0.02164159 0. 0.03607427]\n",
+ " [0. 0.0515639 0.05302105 ... 0.0202692 0. 0. ]\n",
+ " [0. 0.11081323 0.0736592 ... 0.02370018 0. 0.04049514]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.73913043 0.36371761 0.34114053 ... 0.03562945 0.18651726 0.55172414]\n",
+ " [0.74922033 0.36371761 0.32586558 ... 0.03562945 0.1899512 0.55172414]\n",
+ " [0.78132453 0.36952636 0.35539715 ... 0.02692003 0.19555395 0.5066313 ]\n",
+ " ...\n",
+ " [0. 0.00089366 0.0203666 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.20464701 0.02647658 ... 0.02692003 0. 0.12201592]\n",
+ " [0. 0.00089366 0.0203666 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.704623 0.35661305 0.32800407 ... 0.03357086 0.1930779 0.51962865]\n",
+ " [0.67497707 0.34749777 0.30325866 ... 0.03127474 0.19264414 0.49814324]\n",
+ " [0.48684645 0.35464701 0.32545825 ... 0.03182898 0.1749503 0.51564987]\n",
+ " ...\n",
+ " [0. 0.14030384 0.04358452 ... 0.02414885 0. 0.22997347]\n",
+ " [0.06083287 0.08476318 0.09266802 ... 0.02486144 0.02026026 0.12625995]\n",
+ " [0. 0.11228776 0.11059063 ... 0.02731591 0. 0.02360743]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.72559163 0.35357462 0.34725051 ... 0.03285827 0.1964034 0.47718833]\n",
+ " [0.52144561 0.3266756 0.29327902 ... 0.03285827 0.15577444 0.43819629]\n",
+ " [0.62192258 0.35764075 0.3203666 ... 0.03079968 0.19665643 0.4862069 ]\n",
+ " ...\n",
+ " [0. 0.07698838 0.04684318 ... 0.02383215 0. 0.15066313]\n",
+ " [0. 0.10214477 0.11089613 ... 0.02692003 0. 0.03236074]\n",
+ " [0.02577509 0.13789097 0.14775967 ... 0.02913698 0.00133743 0.00371353]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.65360484 0.35886208 0.36829599 ... 0.03072051 0.19351768 0.43607427]\n",
+ " [0.61232801 0.35537682 0.36429056 ... 0.02929533 0.1950238 0.43908046]\n",
+ " [0.5939216 0.35019363 0.34446707 ... 0.03161784 0.17816736 0.45004421]\n",
+ " ...\n",
+ " [0.00496545 0.01843908 0.04365241 ... 0.02528372 0.00083138 0. ]\n",
+ " [0.02721213 0.03696753 0.05424304 ... 0.02639219 0.00261462 0.00495137]\n",
+ " [0. 0.02213286 0.03937542 ... 0.02512536 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.74885342 0.37310098 0.35437882 ... 0.03483769 0.19157781 0.44297082]\n",
+ " [0.4375344 0.37220733 0.31873727 ... 0.03483769 0.19212001 0.45092838]\n",
+ " [0.58301229 0.38114388 0.36761711 ... 0.02850356 0.18904753 0.42175066]\n",
+ " ...\n",
+ " [0. 0.1201966 0.12729124 ... 0.02454473 0. 0. ]\n",
+ " [0. 0.02412869 0.03360489 ... 0.02850356 0. 0. ]\n",
+ " [0. 0.07819482 0.04276986 ... 0.02216944 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.09884908449031964, 0.07944626657520108, 0.11821706722309253, 0.0912182715089516, 0.09458835602841793, 0.06886710438381921, 0.08178222914860407] mean of list_RMSE_SousModele 0.09042405419405801\n",
+ " RMSE resultat vote 0.06331392399873853\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "key 0 value[6] 0.16334850821192484\n",
+ "key 1 value[6] 0.18670485439918758\n",
+ "key 2 value[6] 0.21615527050978905\n",
+ "key 3 value[6] 0.25465259711919624\n",
+ "key 4 value[6] 0.3081780004228992\n",
+ "key 5 value[6] 0.39107838594488675\n",
+ "key 6 value[6] 0.5474158342212556\n",
+ "key 7 value[6] 1.0\n",
+ "key 0 value[2] 0.04498387434911104\n",
+ "key 1 value[2] 0.10432042854944278\n",
+ "key 2 value[2] 0.1707475130839304\n",
+ "key 3 value[2] 0.25229945241832213\n",
+ "key 4 value[2] 0.35231388061950375\n",
+ "key 5 value[2] 0.4849475605485415\n",
+ "key 6 value[2] 0.665838653134753\n",
+ "key 7 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07248385232459577\n",
+ "Current Error err_H 0.06331392399873853\n",
+ "myFeeder.t 7\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.07283289 0.0614053 ... 0.02763262 0. 0.00822281]\n",
+ " [0. 0.12207328 0.09032587 ... 0.0341251 0. 0.08912467]\n",
+ " [0. 0.1165773 0.06904277 ... 0.02992874 0. 0.03660477]\n",
+ " ...\n",
+ " [0. 0.0766756 0.0410387 ... 0.02723674 0. 0.02440318]\n",
+ " [0. 0.0766756 0.0410387 ... 0.02723674 0. 0.02440318]\n",
+ " [0. 0.0766756 0.0410387 ... 0.02723674 0. 0.02440318]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.09862973 0.06171079 ... 0.02501979 0. 0.06525199]\n",
+ " [0. 0.03997617 0.04344874 ... 0.02660333 0. 0. ]\n",
+ " [0. 0.00643432 0.02973523 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.01832994 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01934827 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.13717605 0.03156823 ... 0.02454473 0. 0.23872679]\n",
+ " ...\n",
+ " [0. 0.03485255 0.03156823 ... 0.02454473 0. 0.23872679]\n",
+ " [0. 0.03485255 0.03156823 ... 0.02454473 0. 0.23872679]\n",
+ " [0. 0.00044683 0.02545825 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.07437167 0.07676497 0.0385947 ... 0.02185273 0. 0.2132626 ]\n",
+ " [0.07437167 0.07167113 0.04368635 ... 0.02272367 0. 0.14801061]\n",
+ " [0. 0.01666667 0.00610998 ... 0.02375297 0. 0.14403183]\n",
+ " ...\n",
+ " [0. 0.01179625 0.01293279 ... 0.02232779 0. 0.10106101]\n",
+ " [0. 0.01184093 0.01293279 ... 0.02232779 0. 0.10132626]\n",
+ " [0. 0.01184093 0.01293279 ... 0.02232779 0. 0.10132626]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.06992851 0.05346232 ... 0.0229612 0. 0.13129973]\n",
+ " [0. 0.06769437 0.03665988 ... 0.02256532 0. 0.1198939 ]\n",
+ " [0. 0.01680071 0.02474542 ... 0.02177356 0. 0.01193634]\n",
+ " ...\n",
+ " [0. 0.00049151 0.02535642 ... 0.02177356 0. 0. ]\n",
+ " [0. 0.00049151 0.02535642 ... 0.02177356 0. 0. ]\n",
+ " [0. 0.00049151 0.02535642 ... 0.02177356 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.02472446 0.04534963 ... 0.02385854 0. 0. ]\n",
+ " [0.00528343 0.02695859 0.04494229 ... 0.02422803 0.00087957 0. ]\n",
+ " [0. 0.02886506 0.05125594 ... 0.02781737 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03109301 ... 0.02755344 0. 0. ]\n",
+ " [0. 0.00044683 0.03109301 ... 0.02755344 0. 0. ]\n",
+ " [0. 0.00041704 0.02953157 ... 0.02686725 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.09651475 0.02342159 ... 0.02454473 0. 0. ]\n",
+ " [0.06897817 0.19258266 0.14765784 ... 0.03404592 0.01138623 0. ]\n",
+ " [0. 0.07596068 0.04378819 ... 0.02216944 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00759607 0.02342159 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.0407332 ... 0.0277118 0. 0. ]\n",
+ " [0. 0.00044683 0.0407332 ... 0.0277118 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.05464701 0.04928717 ... 0.02509897 0. 0.04986737]\n",
+ " [0.02091359 0.09593387 0.08197556 ... 0.02391132 0. 0.0198939 ]\n",
+ " [0. 0.05790885 0.04032587 ... 0.02201108 0. 0.01671088]\n",
+ " ...\n",
+ " [0. 0.01568365 0.04266802 ... 0.02019002 0. 0.05066313]\n",
+ " [0. 0.01568365 0.04266802 ... 0.02019002 0. 0.05066313]\n",
+ " [0. 0.01568365 0.04266802 ... 0.02019002 0. 0.05066313]]\n",
+ "list_RMSE_SousModele [0.11178568663006966, 0.07949888136018943, 0.11694763883244787, 0.08653657776144358, 0.09321780131065399, 0.0790937891048099, 0.0976716426856013, 0.0712939348180308] mean of list_RMSE_SousModele 0.09200574406290582\n",
+ " RMSE resultat vote 0.06559129223226406\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 0 value[6] 0.1444116730248319\n",
+ "key 1 value[6] 0.16334850821192484\n",
+ "key 2 value[6] 0.18670485439918758\n",
+ "key 3 value[6] 0.21615527050978905\n",
+ "key 4 value[6] 0.25465259711919624\n",
+ "key 5 value[6] 0.3081780004228992\n",
+ "key 6 value[6] 0.39107838594488675\n",
+ "key 7 value[6] 0.5474158342212556\n",
+ "key 8 value[6] 1.0\n",
+ "key 0 value[2] 0.044125478974989654\n",
+ "key 1 value[2] 0.0951826145943736\n",
+ "key 2 value[2] 0.15057503844798908\n",
+ "key 3 value[2] 0.21615846003120362\n",
+ "key 4 value[2] 0.2931088132923842\n",
+ "key 5 value[2] 0.38907565458649274\n",
+ "key 6 value[2] 0.5103287568330884\n",
+ "key 7 value[2] 0.6873606071688902\n",
+ "key 8 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07162228231305431\n",
+ "Current Error err_H 0.06559129223226406\n",
+ "myFeeder.t 8\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0766756 0.0410387 ... 0.02723674 0. 0.02440318]\n",
+ " [0. 0.0766756 0.0410387 ... 0.02723674 0. 0.02440318]\n",
+ " [0. 0.0766756 0.0410387 ... 0.02723674 0. 0.02440318]\n",
+ " ...\n",
+ " [0.50205467 0.35455764 0.30824847 ... 0.0290578 0.12051328 0.40822281]\n",
+ " [0.4899468 0.35201072 0.31741344 ... 0.03087886 0.10347009 0.42015915]\n",
+ " [0.35789763 0.30759607 0.27790224 ... 0.03721298 0.11001265 0.37400531]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.0041406 0.02851324 ... 0.01847453 0. 0. ]\n",
+ " [0. 0.0041406 0.02817379 ... 0.01847453 0. 0. ]\n",
+ " ...\n",
+ " [0.58944536 0.3536193 0.33021045 ... 0.02792293 0.18867402 0.50044209]\n",
+ " [0.60657983 0.35171284 0.3353021 ... 0.02787015 0.18210736 0.49708223]\n",
+ " [0.60560142 0.32892464 0.32443992 ... 0.02829243 0.18937285 0.46878868]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.54582485e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.54582485e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.54582485e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.12548156e-01 3.61483467e-01 2.79022403e-01 ... 3.48376880e-02\n",
+ " 1.88324598e-01 4.90716180e-01]\n",
+ " [6.12548156e-01 3.61483467e-01 2.79022403e-01 ... 3.48376880e-02\n",
+ " 1.88324598e-01 4.90716180e-01]\n",
+ " [7.76738213e-01 3.08310992e-01 2.99389002e-01 ... 2.61282660e-02\n",
+ " 2.06578710e-01 3.97877984e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01175156 0.01293279 ... 0.02232779 0. 0.10079576]\n",
+ " [0. 0.01018767 0.01283096 ... 0.02169438 0. 0.0867374 ]\n",
+ " [0. 0.01018767 0.01283096 ... 0.02169438 0. 0.0867374 ]\n",
+ " ...\n",
+ " [0.72584847 0.32832887 0.2610998 ... 0.03198733 0.15534068 0.46525199]\n",
+ " [0.72120712 0.2834227 0.2290224 ... 0.02969121 0.13489969 0.42068966]\n",
+ " [0.75443038 0.3626899 0.3287169 ... 0.03380839 0.19705404 0.47320955]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.91510277e-04 2.53564155e-02 ... 2.17735550e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 2.53564155e-02 ... 2.17735550e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 2.53564155e-02 ... 2.17735550e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.23628692e-01 2.87622878e-01 2.12627291e-01 ... 3.07996833e-02\n",
+ " 9.67287186e-02 3.86206897e-01]\n",
+ " [5.62667400e-01 2.82529044e-01 2.01832994e-01 ... 2.96120348e-02\n",
+ " 7.88902946e-02 3.70557029e-01]\n",
+ " [7.64171712e-01 3.28373548e-01 3.22810591e-01 ... 3.88756928e-02\n",
+ " 1.96855232e-01 4.64721485e-01]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.17039023e-04 2.95315682e-02 ... 2.68672473e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.17039023e-04 2.95315682e-02 ... 2.68672473e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.17039023e-04 2.95315682e-02 ... 2.68672473e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.67596160e-01 3.52427763e-01 3.49490835e-01 ... 3.28846661e-02\n",
+ " 1.61021748e-01 4.64898320e-01]\n",
+ " [6.69235003e-01 3.37354781e-01 3.32179226e-01 ... 3.24096068e-02\n",
+ " 1.55431050e-01 4.56056587e-01]\n",
+ " [6.83654375e-01 3.50372356e-01 3.49898167e-01 ... 3.91660069e-02\n",
+ " 1.76745587e-01 4.59239611e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00759607 0.02342159 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00759607 0.02342159 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00759607 0.02342159 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0.70592552 0.34986595 0.34215886 ... 0.03087886 0.18489066 0.4137931 ]\n",
+ " [0.75637498 0.37042002 0.37780041 ... 0.03562945 0.09018616 0.43766578]\n",
+ " [0.70886076 0.36014298 0.35947047 ... 0.03483769 0.19537322 0.43236074]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01568365 0.04266802 ... 0.02019002 0. 0.05066313]\n",
+ " [0. 0.01568365 0.04266802 ... 0.02019002 0. 0.05066313]\n",
+ " [0. 0.01925827 0.04949084 ... 0.01963579 0. 0.03395225]\n",
+ " ...\n",
+ " [0.70410934 0.34405719 0.32291242 ... 0.03372922 0.1392373 0.42758621]\n",
+ " [0.73685562 0.34794459 0.33869654 ... 0.03214568 0.17634195 0.43793103]\n",
+ " [0.67736195 0.32390527 0.26221996 ... 0.03341251 0.14171336 0.35543767]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 3.97827563e-02 ... 2.25917129e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.96469790e-02 ... 2.26972816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.95112016e-02 ... 2.25917129e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.88081698e-01 3.53053321e-01 3.50848608e-01 ... 3.33069411e-02\n",
+ " 9.28851136e-02 4.27055703e-01]\n",
+ " [7.01400355e-01 3.41971999e-01 3.37067210e-01 ... 3.48904724e-02\n",
+ " 9.29333092e-02 4.21927498e-01]\n",
+ " [7.31963554e-01 3.48853143e-01 3.49626612e-01 ... 3.30430193e-02\n",
+ " 1.04982228e-01 4.37135279e-01]]\n",
+ "list_RMSE_SousModele [0.11429703748838639, 0.08219104802288275, 0.12186559759797354, 0.08657238948187716, 0.09559658532596957, 0.09150294902243566, 0.11418596613442898, 0.07613160341779074, 0.06071984321764075] mean of list_RMSE_SousModele 0.09367366885659839\n",
+ " RMSE resultat vote 0.06382766780448922\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 0 value[6] 0.1288294285628787\n",
+ "key 1 value[6] 0.1444116730248319\n",
+ "key 2 value[6] 0.16334850821192484\n",
+ "key 3 value[6] 0.18670485439918758\n",
+ "key 4 value[6] 0.21615527050978905\n",
+ "key 5 value[6] 0.25465259711919624\n",
+ "key 6 value[6] 0.3081780004228992\n",
+ "key 7 value[6] 0.39107838594488675\n",
+ "key 8 value[6] 0.5474158342212556\n",
+ "key 9 value[6] 1.0\n",
+ "key 0 value[2] 0.040772978082836434\n",
+ "key 1 value[2] 0.08330668137830875\n",
+ "key 2 value[2] 0.12904174065160703\n",
+ "key 3 value[2] 0.18232574860393452\n",
+ "key 4 value[2] 0.24354017347527704\n",
+ "key 5 value[2] 0.31737927681029315\n",
+ "key 6 value[2] 0.40714571644580155\n",
+ "key 7 value[2] 0.5290778855815278\n",
+ "key 8 value[2] 0.7023733808782427\n",
+ "key 9 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.07075621403432486\n",
+ "Current Error err_H 0.06382766780448922\n",
+ "myFeeder.t 9\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.31652908 0.30987489 0.29460285 ... 0.03792557 0.09615037 0.3872679 ]\n",
+ " [0.48079252 0.3536193 0.2814664 ... 0.02889945 0.11594072 0.39655172]\n",
+ " [0.48158136 0.38753351 0.3392057 ... 0.03087886 0.14579794 0.42732095]\n",
+ " ...\n",
+ " [0. 0.01260054 0.0101833 ... 0.01749802 0. 0. ]\n",
+ " [0. 0.01121537 0.01120163 ... 0.0175772 0. 0. ]\n",
+ " [0. 0.01255585 0.0101833 ... 0.01749802 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.33486210e-01 3.23056300e-01 3.15071283e-01 ... 2.87674848e-02\n",
+ " 1.89927104e-01 4.57471264e-01]\n",
+ " [7.36843393e-01 3.35597259e-01 3.43448744e-01 ... 3.44154130e-02\n",
+ " 1.86673896e-01 4.63837312e-01]\n",
+ " [7.36843393e-01 3.35597259e-01 3.43448744e-01 ... 3.44154130e-02\n",
+ " 1.86673896e-01 4.63837312e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.36193029e-04 1.46639511e-02 ... 1.77355503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 1.45960625e-02 ... 1.76827659e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 1.46639511e-02 ... 1.77355503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.53439736e-01 3.48972297e-01 3.40122200e-01 ... 2.61282660e-02\n",
+ " 1.96999819e-01 4.50928382e-01]\n",
+ " [7.53439736e-01 3.29758713e-01 2.97352342e-01 ... 2.85035629e-02\n",
+ " 1.91758540e-01 4.27055703e-01]\n",
+ " [7.69216657e-01 3.19481680e-01 2.86150713e-01 ... 3.24623911e-02\n",
+ " 2.03867703e-01 5.06631300e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.54582485e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.54582485e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.54582485e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.75714548 0.35022341 0.33604888 ... 0.03238321 0.19589734 0.48938992]\n",
+ " [0.77415153 0.34374441 0.32250509 ... 0.03087886 0.19573468 0.48832891]\n",
+ " [0.774867 0.32596068 0.31425662 ... 0.0317498 0.19904211 0.47824934]\n",
+ " ...\n",
+ " [0. 0.00205541 0.00529532 ... 0.01813143 0. 0.01193634]\n",
+ " [0. 0.00205541 0.00529532 ... 0.01813143 0. 0.01193634]\n",
+ " [0. 0.00205541 0.00529532 ... 0.01813143 0. 0.01193634]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.65125665e-01 3.40974084e-01 3.21894094e-01 ... 3.91923990e-02\n",
+ " 1.95156335e-01 4.61007958e-01]\n",
+ " [7.58337920e-01 3.45218945e-01 3.22403259e-01 ... 4.03008709e-02\n",
+ " 1.94812941e-01 4.35543767e-01]\n",
+ " [7.43569987e-01 3.58891868e-01 3.22199593e-01 ... 3.50752177e-02\n",
+ " 1.93095970e-01 4.30503979e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 1.84317719e-02 ... 1.86064925e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 1.61914460e-02 ... 1.87648456e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 1.93482688e-02 ... 1.89231987e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.83140708e-01 3.49865952e-01 3.51323829e-01 ... 3.21984693e-02\n",
+ " 1.84830411e-01 4.49336870e-01]\n",
+ " [7.26863572e-01 3.40035746e-01 3.76985743e-01 ... 2.96120348e-02\n",
+ " 1.94397253e-01 4.12908930e-01]\n",
+ " [7.29003853e-01 3.36103664e-01 3.72437203e-01 ... 2.96120348e-02\n",
+ " 1.94361106e-01 4.12732095e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.06404528e-04 2.53903598e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.53224711e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.53903598e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.59677123e-01 3.40929401e-01 3.61507128e-01 ... 2.69200317e-02\n",
+ " 1.93204410e-01 4.32360743e-01]\n",
+ " [7.50687947e-01 3.48078642e-01 3.28920570e-01 ... 2.69200317e-02\n",
+ " 1.92300741e-01 4.32360743e-01]\n",
+ " [7.55641167e-01 3.48078642e-01 3.28920570e-01 ... 2.69200317e-02\n",
+ " 1.93565877e-01 4.35013263e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.34215886e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.34215886e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.34215886e-02 ... 1.82106097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.67198679e-01 3.47631814e-01 3.35641548e-01 ... 3.48376880e-02\n",
+ " 1.60003615e-01 4.01326260e-01]\n",
+ " [7.34268941e-01 3.48704200e-01 3.28818737e-01 ... 3.55502771e-02\n",
+ " 1.51436834e-01 4.02917772e-01]\n",
+ " [7.52485782e-01 3.46872207e-01 3.26782077e-01 ... 3.48376880e-02\n",
+ " 1.80083138e-01 4.14323607e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 1.87667560e-03 2.76985743e-02 ... 1.88440222e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 1.16175156e-03 2.65784114e-02 ... 1.87648456e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.66802444e-02 ... 1.84481394e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[7.39277197e-01 3.46738159e-01 3.40325866e-01 ... 3.28846661e-02\n",
+ " 9.93794807e-02 4.33244916e-01]\n",
+ " [7.15232679e-01 3.53321418e-01 3.59809912e-01 ... 3.30430193e-02\n",
+ " 9.20657871e-02 4.37842617e-01]\n",
+ " [7.11355715e-01 3.59457849e-01 3.62525458e-01 ... 3.42042755e-02\n",
+ " 9.28730646e-02 4.37842617e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 2.70875764e-02 ... 1.88440222e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.66802444e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.70875764e-02 ... 1.89495909e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[6.89671620e-01 3.56613047e-01 3.32892057e-01 ... 3.45209818e-02\n",
+ " 1.85414784e-01 4.11671088e-01]\n",
+ " [7.24802788e-01 3.51474531e-01 3.14460285e-01 ... 3.36500396e-02\n",
+ " 1.76125068e-01 3.92572944e-01]\n",
+ " [7.38414970e-01 3.64566577e-01 3.34012220e-01 ... 3.57878068e-02\n",
+ " 1.88830652e-01 4.23342175e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.40122200e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.38085540e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.02144772e-04 3.02443992e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.12101687906733209, 0.07426611313903203, 0.11589676197810349, 0.07904253186262622, 0.08530146171965178, 0.08346573841816639, 0.11763918230892535, 0.08633085866452506, 0.06425136765363525, 0.06288961245038806] mean of list_RMSE_SousModele 0.08901005072623858\n",
+ " RMSE resultat vote 0.06008641397815921\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 0 value[6] 0.11586412658816056\n",
+ "key 1 value[6] 0.1288294285628787\n",
+ "key 2 value[6] 0.1444116730248319\n",
+ "key 3 value[6] 0.16334850821192484\n",
+ "key 4 value[6] 0.18670485439918758\n",
+ "key 5 value[6] 0.21615527050978905\n",
+ "key 6 value[6] 0.25465259711919624\n",
+ "key 7 value[6] 0.3081780004228992\n",
+ "key 8 value[6] 0.39107838594488675\n",
+ "key 9 value[6] 0.5474158342212556\n",
+ "key 10 value[6] 1.0\n",
+ "key 0 value[2] 0.04027176190198448\n",
+ "key 1 value[2] 0.0783488401765599\n",
+ "key 2 value[2] 0.11846940950339109\n",
+ "key 3 value[2] 0.1643611208442584\n",
+ "key 4 value[2] 0.21586662532988793\n",
+ "key 5 value[2] 0.2758561565716873\n",
+ "key 6 value[2] 0.3465155841426126\n",
+ "key 7 value[2] 0.436685226198922\n",
+ "key 8 value[2] 0.5538276181612931\n",
+ "key 9 value[2] 0.7122226057839325\n",
+ "key 10 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06968923402870829\n",
+ "Current Error err_H 0.06008641397815921\n",
+ "myFeeder.t 10\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.01459606 ... 0.0175772 0. 0. ]\n",
+ " [0. 0.0005064 0.01466395 ... 0.0175772 0. 0. ]\n",
+ " [0. 0.00047662 0.01568228 ... 0.0175772 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00226393 0.01955193 ... 0.01810504 0. 0. ]\n",
+ " [0. 0.00047662 0.01907671 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01446029 ... 0.01794669 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02545825 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01832994 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01832994 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02647658 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.01934827 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02545825 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00205541 0.00519348 ... 0.01797308 0. 0.01193634]\n",
+ " [0. 0.00049151 0.00733198 ... 0.01797308 0. 0. ]\n",
+ " [0. 0.00044683 0.01120163 ... 0.0178939 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0023235 0.01120163 ... 0.01797308 0. 0.01220159]\n",
+ " [0. 0.0023235 0.00529532 ... 0.01781473 0. 0.01220159]\n",
+ " [0. 0.00540661 0.01334012 ... 0.01821061 0. 0.03607427]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.0212831 ... 0.01971496 0. 0. ]\n",
+ " [0. 0.00607685 0.02433809 ... 0.01979414 0. 0. ]\n",
+ " [0. 0.006479 0.02403259 ... 0.01987332 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00049151 0.01537678 ... 0.01828979 0. 0. ]\n",
+ " [0. 0.00049151 0.01812627 ... 0.01884402 0. 0. ]\n",
+ " [0. 0.00044683 0.01527495 ... 0.01844814 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02600136 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00047662 0.02484725 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00047662 0.02661236 ... 0.02148324 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00250223 0.02579769 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.00253202 0.02579769 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00047662 0.02443992 ... 0.01836896 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02342159 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.02342159 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.02647658 ... 0.02058591 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02545825 ... 0.02058591 0. 0. ]\n",
+ " [0. 0.00044683 0.02545825 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02342159 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02637475 ... 0.01828979 0. 0. ]\n",
+ " [0. 0.00205541 0.02566191 ... 0.01947743 0. 0.01193634]\n",
+ " [0. 0.00205541 0.02576375 ... 0.0202692 0. 0.01193634]\n",
+ " ...\n",
+ " [0. 0.02167113 0.03503055 ... 0.01955661 0. 0.0331565 ]\n",
+ " [0. 0.00049151 0.02769857 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00201072 0.02718941 ... 0.01884402 0. 0.01644562]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02776646 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.00047662 0.02681602 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02579769 ... 0.01894959 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.02365207 0.03217923 ... 0.01826339 0. 0. ]\n",
+ " [0. 0.03190349 0.03564155 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.017605 0.03048201 ... 0.01842175 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03268839 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00049151 0.03380855 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00075961 0.03594705 ... 0.01876485 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.01112601 0.03706721 ... 0.01852732 0. 0. ]\n",
+ " [0.02287654 0.0549151 0.07953157 ... 0.02090261 0.01189228 0.05835544]\n",
+ " [0. 0.00889187 0.03706721 ... 0.0192399 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03204345 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.00044683 0.03000679 ... 0.02285563 0. 0. ]\n",
+ " [0. 0.00047662 0.02973523 ... 0.02158881 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00911528 0.03183978 ... 0.02143046 0. 0.09602122]\n",
+ " [0. 0.02838844 0.05037339 ... 0.01963579 0. 0.02511052]\n",
+ " [0. 0.03246947 0.04718262 ... 0.01863288 0. 0.00795756]]\n",
+ "list_RMSE_SousModele [0.07435809303578006, 0.12093720572328338, 0.08161526659971731, 0.08741567702488659, 0.08356203290129799, 0.11202207813640136, 0.08494906031330574, 0.06288729506280866, 0.05933954477397776, 0.054420641018557814] mean of list_RMSE_SousModele 0.08215068945900167\n",
+ " RMSE resultat vote 0.05320058245529627\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 1 value[6] 0.11586412658816056\n",
+ "key 2 value[6] 0.1288294285628787\n",
+ "key 3 value[6] 0.1444116730248319\n",
+ "key 4 value[6] 0.16334850821192484\n",
+ "key 5 value[6] 0.18670485439918758\n",
+ "key 6 value[6] 0.21615527050978905\n",
+ "key 7 value[6] 0.25465259711919624\n",
+ "key 8 value[6] 0.3081780004228992\n",
+ "key 9 value[6] 0.39107838594488675\n",
+ "key 10 value[6] 0.5474158342212556\n",
+ "key 11 value[6] 1.0\n",
+ "key 1 value[2] 0.040329738755718166\n",
+ "key 2 value[2] 0.07714889824909052\n",
+ "key 3 value[2] 0.11893996526765993\n",
+ "key 4 value[2] 0.16523016024719928\n",
+ "key 5 value[2] 0.2182097626780137\n",
+ "key 6 value[2] 0.27904155229291305\n",
+ "key 7 value[2] 0.35312473427053737\n",
+ "key 8 value[2] 0.4441621937728721\n",
+ "key 9 value[2] 0.5573830498356303\n",
+ "key 10 value[2] 0.7205420731938225\n",
+ "key 11 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06819026570385266\n",
+ "Current Error err_H 0.05320058245529627\n",
+ "myFeeder.t 11\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00137027 0.01527495 ... 0.01784112 0. 0. ]\n",
+ " [0. 0.00044683 0.01500339 ... 0.01799947 0. 0. ]\n",
+ " [0. 0.0049151 0.02179226 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02980312 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02973523 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02973523 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00218945 0.00723014 ... 0.01821061 0. 0.01220159]\n",
+ " [0. 0.00044683 0.00784114 ... 0.01844814 0. 0. ]\n",
+ " [0. 0.00049151 0.01598778 ... 0.01844814 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00053619 0.01894094 ... 0.01876485 0. 0. ]\n",
+ " [0. 0.00049151 0.01639511 ... 0.01884402 0. 0. ]\n",
+ " [0. 0.00053619 0.01894094 ... 0.01876485 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00053619 0.01812627 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01527495 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00044683 0.01832994 ... 0.01836896 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00075961 0.02627291 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00080429 0.02668024 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00075961 0.02627291 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02511881 ... 0.01836896 0. 0. ]\n",
+ " [0. 0.00047662 0.02511881 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00253202 0.02627291 ... 0.01847453 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02566191 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02443992 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.02443992 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.02443992 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02647658 ... 0.01979414 0. 0. ]\n",
+ " [0. 0.00044683 0.02647658 ... 0.01979414 0. 0. ]\n",
+ " [0. 0.00044683 0.02647658 ... 0.01979414 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01465594 0.02708758 ... 0.01900238 0. 0.02254642]\n",
+ " [0. 0.01112601 0.0290224 ... 0.01852732 0. 0. ]\n",
+ " [0.03815814 0.01952636 0.03217923 ... 0.01987332 0. 0.01671088]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03523422 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03523422 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03523422 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01757522 0.03095723 ... 0.01799947 0. 0. ]\n",
+ " [0. 0.01611558 0.02837746 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.01974978 0.02817379 ... 0.01805226 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.0269518 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.02701969 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02674813 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.03523422 ... 0.01860649 0. 0. ]\n",
+ " [0. 0.01193029 0.03482688 ... 0.0189232 0. 0. ]\n",
+ " [0.01726289 0.06823056 0.05875764 ... 0.01868567 0. 0.0066313 ]\n",
+ " ...\n",
+ " [0. 0.00049151 0.03564155 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03503055 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00049151 0.03564155 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.02898421 0.05010183 ... 0.01884402 0. 0.00795756]\n",
+ " [0. 0.02606494 0.04684318 ... 0.01863288 0. 0.00795756]\n",
+ " [0. 0.02603515 0.04874406 ... 0.01847453 0. 0.00795756]\n",
+ " ...\n",
+ " [0. 0.00303843 0.03102512 ... 0.01889681 0. 0. ]\n",
+ " [0. 0.00303843 0.03027834 ... 0.01894959 0. 0. ]\n",
+ " [0. 0.00306822 0.03088934 ... 0.01884402 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02640869 ... 0.02338348 0. 0. ]\n",
+ " [0. 0.00047662 0.02613714 ... 0.02385854 0. 0. ]\n",
+ " [0. 0.00047662 0.02545825 ... 0.02359462 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0005064 0.02776646 ... 0.02153603 0. 0. ]\n",
+ " [0. 0.0005064 0.02769857 ... 0.02174716 0. 0. ]\n",
+ " [0. 0.00044683 0.02735913 ... 0.02164159 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.06918806250538334, 0.07759853975654925, 0.08025733933781612, 0.08138220770158854, 0.11071676804008436, 0.0831064279477049, 0.0586843891599534, 0.057337326151689495, 0.04950446296426305, 0.04251759111983908] mean of list_RMSE_SousModele 0.07102931146848715\n",
+ " RMSE resultat vote 0.04595840192499219\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 1 value[6] 0.10497512198316751\n",
+ "key 3 value[6] 0.1288294285628787\n",
+ "key 4 value[6] 0.1444116730248319\n",
+ "key 5 value[6] 0.16334850821192484\n",
+ "key 6 value[6] 0.18670485439918758\n",
+ "key 7 value[6] 0.21615527050978905\n",
+ "key 8 value[6] 0.25465259711919624\n",
+ "key 9 value[6] 0.3081780004228992\n",
+ "key 10 value[6] 0.39107838594488675\n",
+ "key 11 value[6] 0.5474158342212556\n",
+ "key 12 value[6] 1.0\n",
+ "key 1 value[2] 0.039629809815275106\n",
+ "key 3 value[2] 0.07628135423078276\n",
+ "key 4 value[2] 0.1174773440448718\n",
+ "key 5 value[2] 0.16380507154349289\n",
+ "key 6 value[2] 0.2163122878663642\n",
+ "key 7 value[2] 0.2786103818065217\n",
+ "key 8 value[2] 0.3526733944121302\n",
+ "key 9 value[2] 0.4404756867223069\n",
+ "key 10 value[2] 0.5560034812028644\n",
+ "key 11 value[2] 0.7178493631912842\n",
+ "key 12 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06633761038894762\n",
+ "Current Error err_H 0.04595840192499219\n",
+ "myFeeder.t 12\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02980312 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00053619 0.0299389 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03007468 ... 0.01879124 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.01432451 ... 0.01778833 0. 0. ]\n",
+ " [0. 0.00044683 0.01432451 ... 0.01810504 0. 0. ]\n",
+ " [0. 0.00044683 0.01432451 ... 0.01810504 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00053619 0.01894094 ... 0.01876485 0. 0. ]\n",
+ " [0. 0.00049151 0.01639511 ... 0.01884402 0. 0. ]\n",
+ " [0. 0.00058088 0.00814664 ... 0.01868567 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00379803 0.01354379 ... 0.01813143 0. 0.02440318]\n",
+ " [0. 0.00049151 0.00845214 ... 0.01860649 0. 0. ]\n",
+ " [0. 0.00049151 0.00835031 ... 0.01860649 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00071492 0.02484725 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00071492 0.02668024 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00071492 0.02484725 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.01680244 ... 0.01805226 0. 0. ]\n",
+ " [0. 0.00044683 0.01537678 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.01537678 ... 0.01821061 0. 0. ]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02566191 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0005064 0.02613714 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02620502 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02620502 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03523422 ... 0.01908155 0. 0. ]\n",
+ " [0. 0.00044683 0.03238289 ... 0.01939826 0. 0. ]\n",
+ " [0. 0.00044683 0.03136456 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02525458 ... 0.01662708 0. 0. ]\n",
+ " [0. 0.00044683 0.02566191 ... 0.01828979 0. 0. ]\n",
+ " [0. 0.00044683 0.02566191 ... 0.01805226 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02647658 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02593347 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.02579769 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02681602 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.0005064 0.02715547 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.0005064 0.02715547 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03350305 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00053619 0.03441955 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00058088 0.03564155 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00062556 0.03207739 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00058088 0.03472505 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00062556 0.03462322 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00303843 0.03122878 ... 0.01889681 0. 0. ]\n",
+ " [0. 0.00047662 0.03000679 ... 0.01863288 0. 0. ]\n",
+ " [0. 0.0005064 0.0305499 ... 0.01868567 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03258656 ... 0.02069148 0. 0. ]\n",
+ " [0. 0.00053619 0.02837746 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00056598 0.02858113 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02776646 ... 0.02132489 0. 0. ]\n",
+ " [0. 0.00047662 0.02858113 ... 0.02069148 0. 0. ]\n",
+ " [0. 0.00053619 0.02946368 ... 0.01995249 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03944331 ... 0.01947743 0. 0. ]\n",
+ " [0. 0.00044683 0.03781399 ... 0.02042755 0. 0. ]\n",
+ " [0. 0.00044683 0.03781399 ... 0.02042755 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03350305 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.03329939 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.03340122 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03360489 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.00044683 0.03370672 ... 0.01995249 0. 0. ]\n",
+ " [0. 0.00044683 0.03401222 ... 0.02042755 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.07238253823210726, 0.08377936157891423, 0.09184569610563113, 0.08970845174679887, 0.08441027289937704, 0.059771447819728936, 0.06000322740490109, 0.0539648950681397, 0.04709025319291115, 0.054719086236748794] mean of list_RMSE_SousModele 0.06976752302852582\n",
+ " RMSE resultat vote 0.04543888198112536\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 1 value[6] 0.0957523625178593\n",
+ "key 3 value[6] 0.11586412658816056\n",
+ "key 4 value[6] 0.1288294285628787\n",
+ "key 5 value[6] 0.1444116730248319\n",
+ "key 7 value[6] 0.18670485439918758\n",
+ "key 8 value[6] 0.21615527050978905\n",
+ "key 9 value[6] 0.25465259711919624\n",
+ "key 10 value[6] 0.3081780004228992\n",
+ "key 11 value[6] 0.39107838594488675\n",
+ "key 12 value[6] 0.5474158342212556\n",
+ "key 13 value[6] 1.0\n",
+ "key 1 value[2] 0.04040914526321279\n",
+ "key 3 value[2] 0.07762171957512022\n",
+ "key 4 value[2] 0.11952241655808166\n",
+ "key 5 value[2] 0.16688570285244952\n",
+ "key 7 value[2] 0.22027878007053323\n",
+ "key 8 value[2] 0.28298085730896055\n",
+ "key 9 value[2] 0.35483131198578016\n",
+ "key 10 value[2] 0.4441483328001097\n",
+ "key 11 value[2] 0.5577513309259358\n",
+ "key 12 value[2] 0.7167812888983612\n",
+ "key 13 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.0647300158960382\n",
+ "Current Error err_H 0.04543888198112536\n",
+ "myFeeder.t 13\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.01432451 ... 0.01810504 0. 0. ]\n",
+ " [0. 0.00047662 0.0143924 ... 0.01741884 0. 0. ]\n",
+ " [0. 0.00044683 0.0143924 ... 0.01762998 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.16273458 0.11792261 ... 0.03726577 0. 0.27214854]\n",
+ " [0. 0.17551385 0.09891378 ... 0.04228029 0. 0.25800177]\n",
+ " [0. 0.21355377 0.10339443 ... 0.03177619 0. 0.24615385]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.00835031 ... 0.01836896 0. 0. ]\n",
+ " [0. 0.00058088 0.01364562 ... 0.01741884 0. 0. ]\n",
+ " [0. 0.00058088 0.00967413 ... 0.01860649 0. 0. ]\n",
+ " ...\n",
+ " [0.21781325 0.14548704 0.06741344 ... 0.0229612 0. 0.26578249]\n",
+ " [0.1468171 0.1163092 0.05763747 ... 0.02327791 0. 0.22970822]\n",
+ " [0.070776 0.11045576 0.05295316 ... 0.02248614 0. 0.16472149]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02620502 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00041704 0.02498303 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00047662 0.02627291 ... 0.01879124 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02708758 ... 0.02486144 0. 0. ]\n",
+ " [0.30152266 0.17015192 0.10746775 ... 0.02966482 0.09699379 0.21432361]\n",
+ " [0.07362563 0.03946976 0.04677529 ... 0.02575878 0.01936261 0.0351901 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.58655804e-02 ... 1.80522565e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 3.25290438e-02 2.59674134e-02 ... 1.82897862e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 3.93208222e-03 3.23828921e-02 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.22509631e-01 2.10142985e-01 8.80855397e-02 ... 3.01662708e-02\n",
+ " 1.21995301e-01 2.00000000e-01]\n",
+ " [3.69088241e-01 2.32394996e-01 1.32077393e-01 ... 3.01662708e-02\n",
+ " 5.59371046e-02 2.17771883e-01]\n",
+ " [4.42487617e-01 1.82886506e-01 1.10794297e-01 ... 2.94536817e-02\n",
+ " 1.45960600e-01 1.62334218e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02729124 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00047662 0.02749491 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.0005064 0.02749491 ... 0.01821061 0. 0. ]\n",
+ " ...\n",
+ " [0.40331438 0.26085791 0.14725051 ... 0.03483769 0. 0.29195402]\n",
+ " [0.34847429 0.25591302 0.13699932 ... 0.0347849 0. 0.29301503]\n",
+ " [0.21426038 0.20390229 0.12715547 ... 0.03161784 0.00684379 0.20353669]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00067024 0.03533605 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00071492 0.03564155 ... 0.01884402 0. 0. ]\n",
+ " [0. 0.00067024 0.03564155 ... 0.02058591 0. 0. ]\n",
+ " ...\n",
+ " [0.36341956 0.24204647 0.1185336 ... 0.05011876 0. 0.17877984]\n",
+ " [0.32592185 0.26286863 0.14796334 ... 0.04782264 0. 0.21830239]\n",
+ " [0.37259218 0.19021448 0.07739308 ... 0.03008709 0. 0.14509284]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02905635 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.03041412 ... 0.01852732 0. 0. ]\n",
+ " [0. 0.00053619 0.0305499 ... 0.01894959 0. 0. ]\n",
+ " ...\n",
+ " [0.39043601 0.22785225 0.11120163 ... 0.03008709 0. 0.06366048]\n",
+ " [0.22042439 0.21286863 0.10095044 ... 0.03003431 0. 0.23289125]\n",
+ " [0.39093744 0.21435806 0.10672098 ... 0.03204012 0. 0.04668435]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03781399 ... 0.02042755 0. 0. ]\n",
+ " [0. 0.00044683 0.03883232 ... 0.02116653 0. 0. ]\n",
+ " [0. 0.00044683 0.03761032 ... 0.02053312 0. 0. ]\n",
+ " ...\n",
+ " [0.38445545 0.14319333 0.10312288 ... 0.02850356 0. 0.11900973]\n",
+ " [0.39581728 0.19544236 0.10712831 ... 0.03631565 0. 0.20937224]\n",
+ " [0.35703541 0.21105153 0.09633401 ... 0.03626287 0. 0.23076923]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03452138 ... 0.02042755 0. 0. ]\n",
+ " [0. 0.00071492 0.03105906 ... 0.0192399 0. 0. ]\n",
+ " [0. 0.00044683 0.03401222 ... 0.01963579 0. 0. ]\n",
+ " ...\n",
+ " [0.3556962 0.24347632 0.13665988 ... 0.03087886 0. 0.21591512]\n",
+ " [0.3791231 0.21689008 0.11313646 ... 0.03024545 0. 0.13023873]\n",
+ " [0.34439552 0.19575514 0.10763747 ... 0.02882027 0. 0.14190981]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03971487 ... 0.0219583 0. 0. ]\n",
+ " [0. 0.00044683 0.03971487 ... 0.0185801 0. 0. ]\n",
+ " [0. 0.00044683 0.03964698 ... 0.0202692 0. 0. ]\n",
+ " ...\n",
+ " [0.37394973 0.25627048 0.14344874 ... 0.04375825 0. 0.12891247]\n",
+ " [0.37079435 0.24626154 0.13611677 ... 0.03890208 0. 0.14871795]\n",
+ " [0.36120589 0.20759607 0.09979633 ... 0.02966482 0. 0.08152078]]\n",
+ "list_RMSE_SousModele [0.07379127204976892, 0.08418374721654333, 0.09223472021904584, 0.08382910529791864, 0.0637373319067593, 0.07573781759320443, 0.062247982686791256, 0.05733730351013871, 0.0666040603028097, 0.04842025645511599] mean of list_RMSE_SousModele 0.07081235972380961\n",
+ " RMSE resultat vote 0.04963775098677459\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 1 value[6] 0.08787852233288299\n",
+ "key 3 value[6] 0.10497512198316751\n",
+ "key 5 value[6] 0.1288294285628787\n",
+ "key 7 value[6] 0.16334850821192484\n",
+ "key 8 value[6] 0.18670485439918758\n",
+ "key 9 value[6] 0.21615527050978905\n",
+ "key 10 value[6] 0.25465259711919624\n",
+ "key 11 value[6] 0.3081780004228992\n",
+ "key 12 value[6] 0.39107838594488675\n",
+ "key 13 value[6] 0.5474158342212556\n",
+ "key 14 value[6] 1.0\n",
+ "key 1 value[2] 0.03967682582337845\n",
+ "key 3 value[2] 0.07607057998039217\n",
+ "key 5 value[2] 0.11697850680240253\n",
+ "key 7 value[2] 0.16283432785043347\n",
+ "key 8 value[2] 0.2154566908340158\n",
+ "key 9 value[2] 0.2751904077523034\n",
+ "key 10 value[2] 0.34827820182674013\n",
+ "key 11 value[2] 0.4357086076426396\n",
+ "key 12 value[2] 0.5487020605899349\n",
+ "key 13 value[2] 0.7143553591353308\n",
+ "key 14 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06365199697394795\n",
+ "Current Error err_H 0.04963775098677459\n",
+ "myFeeder.t 14\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536, 0.04963775098677459]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.1792672 0.08180584 ... 0.03288467 0. 0.20424403]\n",
+ " [0. 0.1795353 0.07671419 ... 0.02486144 0. 0.19787798]\n",
+ " [0. 0.19088472 0.07053632 ... 0.0260227 0. 0.24350133]\n",
+ " ...\n",
+ " [0. 0.00047662 0.02878479 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.01663272 ... 0.0185801 0. 0. ]\n",
+ " [0. 0.0005064 0.02797013 ... 0.01910794 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[2.85580627e-01 1.77211796e-01 6.82281059e-02 ... 2.43072051e-02\n",
+ " 0.00000000e+00 2.63395225e-01]\n",
+ " [2.12658228e-01 1.30384272e-01 4.94908350e-02 ... 2.12193191e-02\n",
+ " 0.00000000e+00 1.91246684e-01]\n",
+ " [5.31186938e-01 2.73056300e-01 1.86354379e-01 ... 2.73950911e-02\n",
+ " 1.15543105e-01 4.07161804e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 5.49898167e-03 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 9.77596741e-03 ... 1.94774347e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 9.77596741e-03 ... 2.08234363e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.37593102 0.25942806 0.14460285 ... 0.03214568 0.01843485 0.19045093]\n",
+ " [0.36387819 0.19830206 0.12158859 ... 0.03032462 0. 0.17745358]\n",
+ " [0.43436067 0.18981233 0.12942974 ... 0.03072051 0.04536418 0.18567639]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03462322 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00053619 0.02657841 ... 0.01805226 0. 0. ]\n",
+ " [0. 0.00049151 0.03197556 ... 0.01995249 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.23384089 0.21843908 0.11934827 ... 0.03093164 0.00087957 0.20265252]\n",
+ " [0.37069651 0.25320226 0.09714868 ... 0.03404592 0. 0.27108753]\n",
+ " [0.3693145 0.23529937 0.07325187 ... 0.03388757 0. 0.26419098]\n",
+ " ...\n",
+ " [0. 0.0005064 0.02443992 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.02892057 ... 0.0185801 0. 0. ]\n",
+ " [0. 0.00047662 0.02566191 ... 0.01900238 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[3.31150248e-01 2.43789097e-01 1.51323829e-01 ... 2.22486144e-02\n",
+ " 0.00000000e+00 1.69496021e-01]\n",
+ " [3.33736929e-01 1.90929401e-01 9.23625255e-02 ... 3.09580364e-02\n",
+ " 0.00000000e+00 2.08488064e-01]\n",
+ " [5.13281967e-01 2.20732797e-01 1.49592668e-01 ... 2.63657957e-02\n",
+ " 7.43177300e-02 2.22015915e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 7.14924039e-04 3.62525458e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 1.12153709e-02 4.09368635e-02 ... 2.05067300e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 3.61507128e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.37627347 0.19970211 0.10461643 ... 0.02860913 0. 0.05446508]\n",
+ " [0.0259035 0.0914507 0.07223354 ... 0.02866192 0. 0.11671088]\n",
+ " [0. 0.19088472 0.1155465 ... 0.03262074 0. 0.27780725]\n",
+ " ...\n",
+ " [0. 0.00053619 0.0353021 ... 0.02269728 0. 0. ]\n",
+ " [0. 0.00047662 0.03733876 ... 0.02111375 0. 0. ]\n",
+ " [0. 0.00044683 0.03632043 ... 0.02084983 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.24621782 0.07366696 0.05770536 ... 0.02079704 0. 0.01573828]\n",
+ " [0.09395218 0.05677688 0.05790903 ... 0.02512536 0. 0.11016799]\n",
+ " [0.09395218 0.08388442 0.05947047 ... 0.03035102 0. 0.21644562]\n",
+ " ...\n",
+ " [0. 0.00047662 0.03095723 ... 0.02449195 0. 0. ]\n",
+ " [0. 0.00044683 0.03482688 ... 0.02259171 0. 0. ]\n",
+ " [0. 0.00044683 0.032315 ... 0.02259171 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.03802972 0.02100089 0.0398167 ... 0.02114014 0. 0.01591512]\n",
+ " [0.39904605 0.22301162 0.12983707 ... 0.02992874 0. 0.09124668]\n",
+ " [0.36516236 0.23449508 0.13971487 ... 0.03119557 0. 0.166313 ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.0308554 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00044683 0.03004073 ... 0.01741884 0. 0. ]\n",
+ " [0. 0.00044683 0.02718941 ... 0.02058591 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.35920015 0.21483467 0.099389 ... 0.03045658 0. 0.10910698]\n",
+ " [0.43691066 0.29416145 0.20115411 ... 0.03283188 0. 0.31299735]\n",
+ " [0.43192075 0.30598749 0.20183299 ... 0.03394035 0. 0.33138815]\n",
+ " ...\n",
+ " [0. 0.00044683 0.04229464 ... 0.01910794 0. 0. ]\n",
+ " [0. 0.00044683 0.04446707 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00044683 0.04222675 ... 0.01947743 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.27481196 0.17479893 0.0913442 ... 0.02612827 0. 0.07002653]\n",
+ " [0.33973583 0.26854334 0.12810591 ... 0.04615994 0. 0.11671088]\n",
+ " [0.38747019 0.31831993 0.20285132 ... 0.04243864 0.0261341 0.27639257]\n",
+ " ...\n",
+ " [0. 0.00049151 0.0404277 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.04164969 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00049151 0.04083503 ... 0.01868567 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.07357429919010762, 0.08538160674185685, 0.08531056090805654, 0.06850647232554849, 0.07763105220586257, 0.055866897988104375, 0.05623775422907881, 0.07033350151838164, 0.04866654506274673, 0.0513206887732754] mean of list_RMSE_SousModele 0.0672829378943019\n",
+ " RMSE resultat vote 0.04659546922172779\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 1 value[6] 0.08110499228237947\n",
+ "key 3 value[6] 0.0957523625178593\n",
+ "key 7 value[6] 0.1444116730248319\n",
+ "key 8 value[6] 0.16334850821192484\n",
+ "key 9 value[6] 0.18670485439918758\n",
+ "key 10 value[6] 0.21615527050978905\n",
+ "key 11 value[6] 0.25465259711919624\n",
+ "key 12 value[6] 0.3081780004228992\n",
+ "key 13 value[6] 0.39107838594488675\n",
+ "key 14 value[6] 0.5474158342212556\n",
+ "key 15 value[6] 1.0\n",
+ "key 1 value[2] 0.039690572652205845\n",
+ "key 3 value[2] 0.07610511474187936\n",
+ "key 7 value[2] 0.11691367238203988\n",
+ "key 8 value[2] 0.1632550315798591\n",
+ "key 9 value[2] 0.2150771587426072\n",
+ "key 10 value[2] 0.2766316320516597\n",
+ "key 11 value[2] 0.34732196127604925\n",
+ "key 12 value[2] 0.4342796627215436\n",
+ "key 13 value[2] 0.5499217030934566\n",
+ "key 14 value[2] 0.7102526081301894\n",
+ "key 15 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06251489512379994\n",
+ "Current Error err_H 0.04659546922172779\n",
+ "myFeeder.t 15\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536, 0.04963775098677459, 0.04659546922172779]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02973523 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.00047662 0.02837746 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.0005064 0.0293279 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.03421589 ... 0.0202692 0. 0. ]\n",
+ " [0. 0.00044683 0.03370672 ... 0.01995249 0. 0. ]\n",
+ " [0. 0.00044683 0.03462322 ... 0.01931908 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.10840036 0.03370672 ... 0.01860649 0. 0.16923077]\n",
+ " [0. 0.10835567 0.03370672 ... 0.01852732 0. 0.16896552]\n",
+ " [0. 0.1062109 0.03268839 ... 0.01868567 0. 0.16949602]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02600136 ... 0.01879124 0. 0. ]\n",
+ " [0. 0.00047662 0.02600136 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00047662 0.02443992 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.15874293 0.02701969 ... 0.02507258 0. 0.1193634 ]\n",
+ " [0. 0.15874293 0.02701969 ... 0.02507258 0. 0.1193634 ]\n",
+ " [0. 0.15874293 0.02701969 ... 0.02507258 0. 0.1193634 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.03564155 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00120643 0.04195519 ... 0.0212985 0. 0. ]\n",
+ " [0. 0.00053619 0.03676171 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.23114388 0.0395112 ... 0.01821061 0. 0.17692308]\n",
+ " [0. 0.23154602 0.04083503 ... 0.01821061 0. 0.17984085]\n",
+ " [0. 0.23199285 0.04175153 ... 0.01821061 0. 0.17931034]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03788187 ... 0.0212721 0. 0. ]\n",
+ " [0. 0.00044683 0.03781399 ... 0.02285563 0. 0. ]\n",
+ " [0. 0.00044683 0.03041412 ... 0.0236474 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0165922 0.04066531 ... 0.0246503 0. 0.22210433]\n",
+ " [0. 0.0165922 0.04066531 ... 0.0246503 0. 0.22210433]\n",
+ " [0. 0.0165922 0.04066531 ... 0.0246503 0. 0.22210433]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03095723 ... 0.02216944 0. 0. ]\n",
+ " [0. 0.00044683 0.0371351 ... 0.02079704 0. 0. ]\n",
+ " [0. 0.00044683 0.0311609 ... 0.02438638 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.02759674 ... 0.0202692 0. 0. ]\n",
+ " [0. 0.00049151 0.03095723 ... 0.02019002 0. 0. ]\n",
+ " [0. 0.00044683 0.03258656 ... 0.01955661 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00053619 0.03268839 ... 0.02011085 0. 0. ]\n",
+ " [0. 0.00053619 0.03268839 ... 0.02011085 0. 0. ]\n",
+ " [0. 0.00049151 0.03228106 ... 0.02011085 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03937542 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.03971487 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00044683 0.04426341 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]\n",
+ " [0. 0.16753053 0.04467074 ... 0.02623383 0. 0.13757737]\n",
+ " [0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.04175153 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00053619 0.03798371 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00049151 0.04317719 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00049151 0.04185336 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00049151 0.04195519 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00049151 0.04205703 ... 0.01931908 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.03706721 ... 0.02090261 0. 0. ]\n",
+ " [0. 0.00214477 0.03279022 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00210009 0.03391039 ... 0.01931908 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.0165773 0.0407332 ... 0.0253365 0. 0.34350133]\n",
+ " [0. 0.0165773 0.0407332 ... 0.0253365 0. 0.34350133]\n",
+ " [0. 0.01510277 0.0407332 ... 0.02549485 0. 0.30848806]]\n",
+ "list_RMSE_SousModele [0.06513506088352221, 0.07690375306003966, 0.06050164770232898, 0.07478669535726651, 0.05304082820060141, 0.053462892877013175, 0.06065382951358193, 0.048219742888484324, 0.04442979060428345, 0.04223122754003341] mean of list_RMSE_SousModele 0.0579365468627155\n",
+ " RMSE resultat vote 0.03983820166421492\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n",
+ "key 1 value[6] 0.0752353553817643\n",
+ "key 7 value[6] 0.1288294285628787\n",
+ "key 8 value[6] 0.1444116730248319\n",
+ "key 9 value[6] 0.16334850821192484\n",
+ "key 10 value[6] 0.18670485439918758\n",
+ "key 11 value[6] 0.21615527050978905\n",
+ "key 12 value[6] 0.25465259711919624\n",
+ "key 13 value[6] 0.3081780004228992\n",
+ "key 14 value[6] 0.39107838594488675\n",
+ "key 15 value[6] 0.5474158342212556\n",
+ "key 16 value[6] 1.0\n",
+ "key 1 value[2] 0.04057300926516191\n",
+ "key 7 value[2] 0.07778641425198637\n",
+ "key 8 value[2] 0.11997348108189705\n",
+ "key 9 value[2] 0.16655647365237794\n",
+ "key 10 value[2] 0.22086636713220287\n",
+ "key 11 value[2] 0.2817972630862373\n",
+ "key 12 value[2] 0.3540580001222705\n",
+ "key 13 value[2] 0.4436794114741592\n",
+ "key 14 value[2] 0.5580391463019391\n",
+ "key 15 value[2] 0.719538937462349\n",
+ "key 16 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.061097601782575874\n",
+ "Current Error err_H 0.03983820166421492\n",
+ "myFeeder.t 16\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536, 0.04963775098677459, 0.04659546922172779, 0.03983820166421492]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " [0. 0.18418231 0.04867617 ... 0.02359462 0. 0.27780725]\n",
+ " ...\n",
+ " [0.40069712 0.35963658 0.31228785 ... 0.02897862 0.14558708 0.49425287]\n",
+ " [0.40069712 0.35963658 0.31228785 ... 0.02897862 0.14558708 0.49425287]\n",
+ " [0.40069712 0.35963658 0.31228785 ... 0.02897862 0.14558708 0.49425287]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.15853441 0.02708758 ... 0.02507258 0. 0.11794872]\n",
+ " [0. 0.15853441 0.0269518 ... 0.02501979 0. 0.11794872]\n",
+ " [0. 0.15823652 0.0269518 ... 0.02491423 0. 0.11653404]\n",
+ " ...\n",
+ " [0.44209625 0.369437 0.33340122 ... 0.04206915 0.16682933 0.44686118]\n",
+ " [0.44209625 0.369437 0.33340122 ... 0.04206915 0.16682933 0.44686118]\n",
+ " [0.46375589 0.37003277 0.33021045 ... 0.03531275 0.17251642 0.44739169]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.23154602 0.04185336 ... 0.01821061 0. 0.17267905]\n",
+ " [0. 0.23154602 0.04185336 ... 0.01821061 0. 0.17267905]\n",
+ " [0. 0.2310992 0.04093686 ... 0.01821061 0. 0.17320955]\n",
+ " ...\n",
+ " [0.48653458 0.35169794 0.33594705 ... 0.03499604 0.19215615 0.41856764]\n",
+ " [0.48653458 0.35169794 0.33594705 ... 0.03499604 0.19215615 0.41856764]\n",
+ " [0.48653458 0.35169794 0.33594705 ... 0.03499604 0.19215615 0.41856764]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0165922 0.04066531 ... 0.0246503 0. 0.22210433]\n",
+ " [0. 0.01656241 0.03910387 ... 0.02459752 0. 0.22687887]\n",
+ " [0. 0.01656241 0.03910387 ... 0.02459752 0. 0.22687887]\n",
+ " ...\n",
+ " [0.40042806 0.36121537 0.35580448 ... 0.03346529 0.14526176 0.4351901 ]\n",
+ " [0.40042806 0.36121537 0.35580448 ... 0.03346529 0.14526176 0.4351901 ]\n",
+ " [0.40097841 0.36127495 0.35634759 ... 0.03388757 0.14533406 0.4351901 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " [0. 0.01626452 0.03625255 ... 0.02375297 0. 0.25923961]\n",
+ " ...\n",
+ " [0.38188712 0.33848674 0.34120842 ... 0.03462655 0.12169408 0.41061008]\n",
+ " [0.38188712 0.33848674 0.34120842 ... 0.03462655 0.12169408 0.41061008]\n",
+ " [0.38425977 0.34206136 0.34405974 ... 0.03462655 0.12165793 0.41662246]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.03228106 ... 0.02011085 0. 0. ]\n",
+ " [0. 0.00049151 0.03228106 ... 0.02011085 0. 0. ]\n",
+ " [0. 0.00053619 0.03268839 ... 0.02011085 0. 0. ]\n",
+ " ...\n",
+ " [0.38350761 0.35862377 0.34704684 ... 0.04251781 0.16282306 0.45941645]\n",
+ " [0.38350761 0.35862377 0.34704684 ... 0.04251781 0.16282306 0.45941645]\n",
+ " [0.4023115 0.35889187 0.34633401 ... 0.04259699 0.17202241 0.4596817 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]\n",
+ " [0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]\n",
+ " [0. 0.17679476 0.04555329 ... 0.02665611 0. 0.14394341]\n",
+ " ...\n",
+ " [0.37485477 0.34864462 0.34453496 ... 0.03177619 0.1281282 0.47975243]\n",
+ " [0.37485477 0.34864462 0.34453496 ... 0.03177619 0.1281282 0.47975243]\n",
+ " [0.37392527 0.3484361 0.34494229 ... 0.03198733 0.13439364 0.4795756 ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00049151 0.0413442 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00049151 0.0413442 ... 0.01931908 0. 0. ]\n",
+ " [0. 0.00049151 0.04185336 ... 0.01931908 0. 0. ]\n",
+ " ...\n",
+ " [0.38783709 0.34941912 0.34653768 ... 0.03515439 0.15711187 0.47320955]\n",
+ " [0.38844249 0.34946381 0.34796334 ... 0.03515439 0.15707573 0.47320955]\n",
+ " [0.38798386 0.35344057 0.34399185 ... 0.03871734 0.1482017 0.47559682]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.01662198 0.0407332 ... 0.02549485 0. 0.33793103]\n",
+ " [0. 0.0165773 0.0407332 ... 0.02549485 0. 0.33766578]\n",
+ " [0. 0.0165773 0.0407332 ... 0.02541568 0. 0.34084881]\n",
+ " ...\n",
+ " [0.35545771 0.35013405 0.35376782 ... 0.0277118 0.11941081 0.47188329]\n",
+ " [0.35676023 0.34812332 0.35855397 ... 0.02763262 0.12750768 0.47241379]\n",
+ " [0.35916346 0.34517426 0.3392057 ... 0.02763262 0.11919393 0.47267905]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.03479297 0.0377461 ... 0.01921351 0. 0.29513705]\n",
+ " [0. 0.02993744 0.03842498 ... 0.01921351 0. 0.26843501]\n",
+ " [0. 0.02993744 0.03842498 ... 0.01921351 0. 0.26843501]\n",
+ " ...\n",
+ " [0.37309362 0.34331248 0.32613714 ... 0.02955925 0.17898669 0.52590628]\n",
+ " [0.37309362 0.34331248 0.32613714 ... 0.02955925 0.17898669 0.52590628]\n",
+ " [0.37907418 0.34331248 0.32260692 ... 0.02950647 0.17891439 0.52838196]]\n",
+ "list_RMSE_SousModele [0.07054819272510317, 0.07085544345128844, 0.100644780871035, 0.07649758815281328, 0.07580005152596367, 0.09191452250831013, 0.07727621737184309, 0.07946217463938195, 0.07076410566665609, 0.0570308066245192] mean of list_RMSE_SousModele 0.07707938835369141\n",
+ " RMSE resultat vote 0.060894731847603654\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 1 value[6] 0.0701133935324861\n",
+ "key 8 value[6] 0.1288294285628787\n",
+ "key 9 value[6] 0.1444116730248319\n",
+ "key 10 value[6] 0.16334850821192484\n",
+ "key 11 value[6] 0.18670485439918758\n",
+ "key 12 value[6] 0.21615527050978905\n",
+ "key 13 value[6] 0.25465259711919624\n",
+ "key 14 value[6] 0.3081780004228992\n",
+ "key 15 value[6] 0.39107838594488675\n",
+ "key 16 value[6] 0.5474158342212556\n",
+ "key 17 value[6] 1.0\n",
+ "key 1 value[2] 0.03954093492437874\n",
+ "key 8 value[2] 0.07504892468099995\n",
+ "key 9 value[2] 0.11484694649185888\n",
+ "key 10 value[2] 0.16126021707735014\n",
+ "key 11 value[2] 0.21211742811555354\n",
+ "key 12 value[2] 0.27129388082423983\n",
+ "key 13 value[2] 0.3428582746327872\n",
+ "key 14 value[2] 0.43020500693991615\n",
+ "key 15 value[2] 0.5456164358635724\n",
+ "key 16 value[2] 0.7110777899299824\n",
+ "key 17 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.06108566825698928\n",
+ "Current Error err_H 0.060894731847603654\n",
+ "myFeeder.t 17\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536, 0.04963775098677459, 0.04659546922172779, 0.03983820166421492, 0.060894731847603654]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.72818443e-01 3.55942806e-01 2.98031229e-01 ... 2.86619161e-02\n",
+ " 1.57141997e-01 4.98320071e-01]\n",
+ " [4.78040726e-01 3.53410783e-01 2.84657162e-01 ... 2.85563473e-02\n",
+ " 1.56744382e-01 5.00442087e-01]\n",
+ " [4.74188222e-01 3.54959786e-01 2.94976239e-01 ... 2.86619161e-02\n",
+ " 1.57141997e-01 4.98320071e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 1.64969450e-02 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 1.64969450e-02 ... 1.85273159e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 1.83299389e-02 ... 1.84745315e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.63645814e-01 3.70539172e-01 3.31228785e-01 ... 3.53127474e-02\n",
+ " 1.72697150e-01 4.48099027e-01]\n",
+ " [4.80535682e-01 3.65564492e-01 3.27359131e-01 ... 3.56294537e-02\n",
+ " 1.72829689e-01 4.52343059e-01]\n",
+ " [5.37760656e-01 3.66815609e-01 3.27494908e-01 ... 3.58933756e-02\n",
+ " 1.79143322e-01 4.54465075e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.36193029e-04 2.89884589e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 2.89205703e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 2.89205703e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.10297805e-01 3.61006851e-01 3.56076035e-01 ... 3.35180786e-02\n",
+ " 1.45587084e-01 4.35013263e-01]\n",
+ " [4.46743717e-01 3.62228180e-01 3.57230143e-01 ... 3.33597255e-02\n",
+ " 1.58093861e-01 4.35190097e-01]\n",
+ " [6.48590473e-01 3.73935061e-01 3.63340122e-01 ... 3.49432568e-02\n",
+ " 1.77866137e-01 4.43147657e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 3.07535642e-02 ... 1.99524941e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.08214528e-02 ... 2.04275534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.21113374e-02 ... 2.04275534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.08512200e-01 3.43312481e-01 3.57094365e-01 ... 3.41514912e-02\n",
+ " 1.44454485e-01 4.10610080e-01]\n",
+ " [5.17348499e-01 3.40601728e-01 3.38424983e-01 ... 3.34125099e-02\n",
+ " 1.63805048e-01 4.12908930e-01]\n",
+ " [5.31462117e-01 3.43521001e-01 3.40665309e-01 ... 3.37820005e-02\n",
+ " 1.48816194e-01 4.12378426e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.73387644e-02 ... 2.22750066e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.04616429e-02 ... 2.15888097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.07331976e-02 ... 2.13776722e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.12217942e-01 3.49955317e-01 3.38391039e-01 ... 3.52335709e-02\n",
+ " 1.81691668e-01 4.48806366e-01]\n",
+ " [4.48651624e-01 3.50938338e-01 3.47250509e-01 ... 3.53127474e-02\n",
+ " 1.81962769e-01 4.58885942e-01]\n",
+ " [5.39112090e-01 3.50536193e-01 3.30244399e-01 ... 3.48376880e-02\n",
+ " 1.75293692e-01 4.53050398e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.91510277e-04 3.16700611e-02 ... 1.90815519e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.48268839e-02 ... 1.91607284e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 3.27902240e-02 ... 1.90815519e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[3.93273405e-01 3.46201966e-01 3.47114732e-01 ... 3.19345474e-02\n",
+ " 1.27489608e-01 4.75685234e-01]\n",
+ " [5.21604599e-01 3.39469765e-01 3.40529532e-01 ... 3.81103193e-02\n",
+ " 1.54792457e-01 4.65959328e-01]\n",
+ " [5.42481502e-01 3.49180816e-01 3.41819416e-01 ... 3.40987068e-02\n",
+ " 1.66829327e-01 4.63306808e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 4.42634080e-02 ... 1.95302191e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.31093007e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.18873048e-02 ... 1.92662972e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.29884425e-01 3.56344951e-01 3.48167006e-01 ... 3.68171021e-02\n",
+ " 1.24399060e-01 4.73474801e-01]\n",
+ " [6.14180884e-01 3.51027703e-01 3.43177189e-01 ... 3.48376880e-02\n",
+ " 1.24236400e-01 4.72413793e-01]\n",
+ " [6.39552376e-01 3.35031278e-01 3.26680244e-01 ... 3.47585115e-02\n",
+ " 1.24615941e-01 4.73740053e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 4.27698574e-02 ... 1.83689628e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.27698574e-02 ... 1.86856690e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.33808554e-02 ... 1.86856690e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[3.80737479e-01 3.45844504e-01 3.27800407e-01 ... 2.80285036e-02\n",
+ " 1.39454184e-01 4.78249337e-01]\n",
+ " [4.26417171e-01 3.64075067e-01 3.19857434e-01 ... 3.02454473e-02\n",
+ " 1.64594253e-01 5.12466844e-01]\n",
+ " [5.83966245e-01 3.33869526e-01 3.14562118e-01 ... 2.62866192e-02\n",
+ " 1.40556660e-01 4.64986737e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.98167006e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.03258656e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.11405295e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.17856051e-01 3.42269884e-01 3.25661914e-01 ... 3.00870942e-02\n",
+ " 1.78492680e-01 5.25198939e-01]\n",
+ " [5.78438207e-01 3.46559428e-01 3.34962661e-01 ... 2.87147004e-02\n",
+ " 1.65744924e-01 5.13704686e-01]\n",
+ " [5.85543937e-01 3.53827822e-01 3.32993890e-01 ... 2.83452098e-02\n",
+ " 1.58889090e-01 5.20070734e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 3.83570944e-02 ... 1.96885722e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.63204345e-02 ... 1.94774347e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.72708758e-02 ... 1.95302191e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.82572005e-01 3.29669348e-01 3.59063136e-01 ... 3.08788599e-02\n",
+ " 1.70233147e-01 5.25464191e-01]\n",
+ " [4.13006788e-01 3.29669348e-01 3.45621181e-01 ... 3.34125099e-02\n",
+ " 1.40809687e-01 5.23607427e-01]\n",
+ " [5.64135021e-01 3.45040214e-01 3.42973523e-01 ... 3.08788599e-02\n",
+ " 1.31574191e-01 5.19363395e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.36193029e-04 3.51323829e-02 ... 1.92399050e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 3.56415479e-02 ... 1.91607284e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.78818737e-02 ... 1.90815519e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.0634462955129073, 0.06179048808901572, 0.061152501012590114, 0.06044073063835678, 0.07254141175453, 0.06106228828206411, 0.06231934217527529, 0.058366134983545176, 0.04757304058235922, 0.047341463979297144] mean of list_RMSE_SousModele 0.05960336970099409\n",
+ " RMSE resultat vote 0.044176494409285595\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle KNN\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "key 1 value[6] 0.0656141137913348\n",
+ "key 8 value[6] 0.11586412658816056\n",
+ "key 10 value[6] 0.1444116730248319\n",
+ "key 11 value[6] 0.16334850821192484\n",
+ "key 12 value[6] 0.18670485439918758\n",
+ "key 13 value[6] 0.21615527050978905\n",
+ "key 14 value[6] 0.25465259711919624\n",
+ "key 15 value[6] 0.3081780004228992\n",
+ "key 16 value[6] 0.39107838594488675\n",
+ "key 17 value[6] 0.5474158342212556\n",
+ "key 18 value[6] 1.0\n",
+ "key 1 value[2] 0.040755422872325676\n",
+ "key 8 value[2] 0.07791942384583156\n",
+ "key 10 value[2] 0.1193809995089374\n",
+ "key 11 value[2] 0.1658260358301378\n",
+ "key 12 value[2] 0.21856644171652162\n",
+ "key 13 value[2] 0.2802456486749389\n",
+ "key 14 value[2] 0.35246931136145176\n",
+ "key 15 value[2] 0.4418657653067841\n",
+ "key 16 value[2] 0.5573430614308316\n",
+ "key 17 value[2] 0.717469394245067\n",
+ "key 18 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.060146269709894624\n",
+ "Current Error err_H 0.044176494409285595\n",
+ "myFeeder.t 18\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536, 0.04963775098677459, 0.04659546922172779, 0.03983820166421492, 0.060894731847603654, 0.044176494409285595]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 2.12491514e-02 ... 1.87384534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.08418194e-02 ... 1.84745315e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.17039023e-04 1.70400543e-02 ... 1.84217472e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.86069834e-01 3.51921358e-01 3.47386286e-01 ... 2.78701504e-02\n",
+ " 1.74504488e-01 4.79929266e-01]\n",
+ " [5.86155445e-01 3.52874590e-01 3.53496266e-01 ... 2.79757192e-02\n",
+ " 1.87517320e-01 4.78337754e-01]\n",
+ " [5.84406531e-01 3.52785225e-01 3.52002716e-01 ... 2.83979942e-02\n",
+ " 1.86987168e-01 4.77630416e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.02844535 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.0005064 0.02851324 ... 0.01921351 0. 0. ]\n",
+ " [0. 0.0005064 0.02885268 ... 0.01921351 0. 0. ]\n",
+ " ...\n",
+ " [0.43881857 0.3497468 0.34487441 ... 0.03394035 0.093596 0.40212202]\n",
+ " [0.45931633 0.35108728 0.34480652 ... 0.03404592 0.09352371 0.40229885]\n",
+ " [0.47064147 0.34673816 0.3424983 ... 0.03372922 0.0930538 0.40707339]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 3.10251188e-02 ... 1.98997097e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 3.34012220e-02 ... 2.05859066e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 3.68635438e-02 ... 2.05859066e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.47312420e-01 3.75364909e-01 3.22946368e-01 ... 4.75587226e-02\n",
+ " 1.63853244e-01 4.29531388e-01]\n",
+ " [5.85971993e-01 3.77360739e-01 3.33469111e-01 ... 4.75587226e-02\n",
+ " 1.76408217e-01 4.36604775e-01]\n",
+ " [6.05148902e-01 3.81173667e-01 3.42498303e-01 ... 3.55238849e-02\n",
+ " 1.88517381e-01 4.40848806e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.0407332 ... 0.02137767 0. 0. ]\n",
+ " [0. 0.00047662 0.03272234 ... 0.02370018 0. 0. ]\n",
+ " [0. 0.00044683 0.03340122 ... 0.02248614 0. 0. ]\n",
+ " ...\n",
+ " [0.18293891 0.19133155 0.18065173 ... 0.03198733 0.06702813 0.31335102]\n",
+ " [0.20964961 0.21027703 0.20074678 ... 0.03225125 0.0796554 0.32484527]\n",
+ " [0.28630832 0.26273458 0.2496945 ... 0.03314859 0.10503042 0.35932803]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 3.96469790e-02 ... 1.90551597e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.34487441e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.43991853e-02 ... 1.94774347e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [4.80132086e-01 3.39469765e-01 3.13577733e-01 ... 3.38875693e-02\n",
+ " 1.02644738e-01 4.30769231e-01]\n",
+ " [4.40836544e-01 3.43997617e-01 3.32993890e-01 ... 3.39931380e-02\n",
+ " 1.21151877e-01 4.48629531e-01]\n",
+ " [3.86705803e-01 3.49270182e-01 3.47522064e-01 ... 3.43098443e-02\n",
+ " 1.31140430e-01 4.66666667e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 4.36863544e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.21588595e-02 ... 1.86856690e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 4.39918534e-02 ... 1.86856690e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.15630160e-01 3.61751564e-01 3.11201629e-01 ... 3.43626287e-02\n",
+ " 1.26730526e-01 4.84084881e-01]\n",
+ " [5.44945881e-01 3.60634495e-01 3.05397149e-01 ... 3.44418052e-02\n",
+ " 1.17404663e-01 4.84084881e-01]\n",
+ " [4.67932489e-01 3.59785523e-01 3.13951120e-01 ... 3.44418052e-02\n",
+ " 1.39634918e-01 4.82758621e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 3.91038697e-02 ... 1.97149644e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 3.83910387e-02 ... 2.04275534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 3.77800407e-02 ... 1.97149644e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [6.56558430e-01 3.62153709e-01 3.54684318e-01 ... 3.15122724e-02\n",
+ " 1.64937647e-01 5.28381963e-01]\n",
+ " [4.98660796e-01 2.88963360e-01 2.75458248e-01 ... 2.81868567e-02\n",
+ " 1.26043738e-01 4.14058355e-01]\n",
+ " [5.58007705e-01 3.60768543e-01 3.33401222e-01 ... 2.99287411e-02\n",
+ " 1.67739020e-01 5.06100796e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.76616026e-04 3.65919891e-02 ... 1.95302191e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.74745418e-02 ... 1.92662972e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 3.72708758e-02 ... 1.91607284e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [4.82553660e-01 3.63002681e-01 3.03326544e-01 ... 2.94536817e-02\n",
+ " 1.47575155e-01 5.25906278e-01]\n",
+ " [4.82810493e-01 3.62615430e-01 3.04752206e-01 ... 2.95064661e-02\n",
+ " 1.41092837e-01 5.24491600e-01]\n",
+ " [5.03002507e-01 3.54185284e-01 2.96266124e-01 ... 2.95592505e-02\n",
+ " 1.37008254e-01 5.21308576e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 4.46827525e-04 4.02240326e-02 ... 1.90815519e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 3.95112016e-02 ... 1.92399050e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.72708758e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [4.12016144e-01 3.43655049e-01 3.11812627e-01 ... 3.16706255e-02\n",
+ " 1.59732514e-01 5.38461538e-01]\n",
+ " [4.33773620e-01 3.44906166e-01 3.15173116e-01 ... 3.14330958e-02\n",
+ " 1.69889752e-01 5.39257294e-01]\n",
+ " [4.51935425e-01 3.52993744e-01 3.29124236e-01 ... 3.17498021e-02\n",
+ " 1.78944515e-01 5.48541114e-01]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0.00000000e+00 5.06404528e-04 3.57094365e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.59809912e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.66598778e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " ...\n",
+ " [5.17006054e-01 3.56687519e-01 3.19076714e-01 ... 2.87147004e-02\n",
+ " 1.31140430e-01 5.33863837e-01]\n",
+ " [5.06121201e-01 3.55793864e-01 3.15003394e-01 ... 2.84507786e-02\n",
+ " 1.29152359e-01 5.29089302e-01]\n",
+ " [4.93695346e-01 3.55436402e-01 3.20027155e-01 ... 2.84507786e-02\n",
+ " 1.34212904e-01 5.32272325e-01]]\n",
+ "list_RMSE_SousModele [0.07005809004326741, 0.06706070228249626, 0.06757106425445485, 0.06842351091427931, 0.07310672912315051, 0.07848568711012492, 0.07084912809120027, 0.0566023801581399, 0.058851554477125846, 0.05138410398329418] mean of list_RMSE_SousModele 0.06623929504375334\n",
+ " RMSE resultat vote 0.05113756750316481\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle R_Forest\n",
+ "key 1 value[6] 0.06163692269875859\n",
+ "key 8 value[6] 0.10497512198316751\n",
+ "key 10 value[6] 0.1288294285628787\n",
+ "key 11 value[6] 0.1444116730248319\n",
+ "key 13 value[6] 0.18670485439918758\n",
+ "key 14 value[6] 0.21615527050978905\n",
+ "key 15 value[6] 0.25465259711919624\n",
+ "key 16 value[6] 0.3081780004228992\n",
+ "key 17 value[6] 0.39107838594488675\n",
+ "key 18 value[6] 0.5474158342212556\n",
+ "key 19 value[6] 1.0\n",
+ "key 1 value[2] 0.03989996241154676\n",
+ "key 8 value[2] 0.07634992684879684\n",
+ "key 10 value[2] 0.11695369800505323\n",
+ "key 11 value[2] 0.16247547397035103\n",
+ "key 13 value[2] 0.21424867229813513\n",
+ "key 14 value[2] 0.27469674181899106\n",
+ "key 15 value[2] 0.3475368439178264\n",
+ "key 16 value[2] 0.43671801203861993\n",
+ "key 17 value[2] 0.5499508758437817\n",
+ "key 18 value[2] 0.7137100013970308\n",
+ "key 19 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.0596721274884878\n",
+ "Current Error err_H 0.05113756750316481\n",
+ "myFeeder.t 19\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536, 0.04963775098677459, 0.04659546922172779, 0.03983820166421492, 0.060894731847603654, 0.044176494409285595, 0.05113756750316481]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.90509387e-01 3.47453083e-01 3.36591989e-01 ... 2.79757192e-02\n",
+ " 1.88529429e-01 4.87002653e-01]\n",
+ " [5.88026662e-01 3.27852249e-01 3.02036660e-01 ... 2.72895223e-02\n",
+ " 1.83709862e-01 4.68435013e-01]\n",
+ " [5.82975601e-01 3.14060173e-01 2.90088255e-01 ... 2.75534442e-02\n",
+ " 1.89288511e-01 4.57648099e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.53903598e-02 ... 1.88440222e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 2.35573659e-02 ... 1.89495909e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 1.41887305e-02 ... 1.78939034e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.58038280e-01 3.50580876e-01 3.56143924e-01 ... 3.33069411e-02\n",
+ " 9.41863968e-02 4.34305924e-01]\n",
+ " [4.41912799e-01 3.50848972e-01 3.27698574e-01 ... 3.17234099e-02\n",
+ " 9.14874390e-02 4.35897436e-01]\n",
+ " [4.39870360e-01 3.49865952e-01 3.16836388e-01 ... 3.39931380e-02\n",
+ " 1.10693415e-01 4.40848806e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.44399185e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 2.44399185e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.86833482e-03 2.86490156e-02 ... 1.88968065e-02\n",
+ " 0.00000000e+00 3.00618921e-03]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.75343974e-01 3.80190646e-01 3.39511202e-01 ... 3.54711006e-02\n",
+ " 1.90240376e-01 4.39964633e-01]\n",
+ " [4.65896166e-01 3.69645517e-01 3.27359131e-01 ... 3.51016099e-02\n",
+ " 1.51141635e-01 4.45092838e-01]\n",
+ " [4.01284168e-01 3.66070897e-01 3.23421589e-01 ... 3.46793349e-02\n",
+ " 1.32020001e-01 4.48629531e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.36193029e-04 3.69993211e-02 ... 2.02692003e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.36193029e-04 3.26544467e-02 ... 2.06914753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.65981531e-04 3.31975560e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.87311197e-01 3.45546619e-01 3.37746096e-01 ... 3.42042755e-02\n",
+ " 1.47454666e-01 4.14854111e-01]\n",
+ " [4.06164007e-01 3.58802502e-01 3.38289206e-01 ... 3.41514912e-02\n",
+ " 1.34321345e-01 4.13085765e-01]\n",
+ " [3.81312297e-01 3.60887697e-01 3.42226748e-01 ... 3.42570599e-02\n",
+ " 1.41888066e-01 4.13969938e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 3.17040054e-02 ... 2.29612035e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.01425662e-02 ... 1.94774347e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.17718941e-02 ... 2.07970441e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.12902831e-01 3.49210605e-01 3.44738629e-01 ... 3.43098443e-02\n",
+ " 1.32501958e-01 4.67374005e-01]\n",
+ " [3.92747508e-01 3.50431933e-01 3.23964698e-01 ... 3.40459224e-02\n",
+ " 1.14091210e-01 4.67904509e-01]\n",
+ " [4.15030881e-01 3.54810843e-01 3.19212492e-01 ... 3.38347849e-02\n",
+ " 1.13175492e-01 4.64544651e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.76616026e-04 4.26340801e-02 ... 1.95830034e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.76103191e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.65919891e-02 ... 1.94246503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.32397725e-01 3.63092046e-01 3.20162933e-01 ... 2.93745051e-02\n",
+ " 1.37538406e-01 4.99469496e-01]\n",
+ " [4.01045680e-01 3.10634495e-01 2.28513238e-01 ... 4.79809976e-02\n",
+ " 7.47876378e-02 4.55968170e-01]\n",
+ " [4.08438819e-01 3.14253798e-01 2.46028513e-01 ... 3.68962787e-02\n",
+ " 7.69202964e-02 4.69230769e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.93279022e-02 ... 1.97941409e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 3.79803396e-03 3.17718941e-02 ... 2.04275534e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [3.39020363e-02 3.17694370e-02 3.73727088e-02 ... 2.05067300e-02\n",
+ " 0.00000000e+00 1.16710875e-02]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[5.71662692e-01 3.62853738e-01 3.13170401e-01 ... 3.09844286e-02\n",
+ " 1.68781252e-01 5.40406720e-01]\n",
+ " [4.51672476e-01 3.71105153e-01 3.20366599e-01 ... 3.14067036e-02\n",
+ " 1.86059401e-01 5.48541114e-01]\n",
+ " [4.72952975e-01 3.62645219e-01 3.14052953e-01 ... 4.67141726e-02\n",
+ " 1.79420447e-01 5.46772767e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 5.06404528e-04 2.71554650e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.11608961e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 5.06404528e-04 3.00746775e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.90772335e-01 3.49329759e-01 3.27291242e-01 ... 3.17498021e-02\n",
+ " 1.79540936e-01 5.47480106e-01]\n",
+ " [1.12126215e-01 2.09740840e-01 1.75661914e-01 ... 2.70783848e-02\n",
+ " 3.70865715e-02 3.09549072e-01]\n",
+ " [0.00000000e+00 1.20062556e-01 1.16395112e-01 ... 3.05621536e-02\n",
+ " 0.00000000e+00 6.02122016e-02]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 2.94297352e-02 ... 1.97149644e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.91510277e-04 3.38085540e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 2.05987489e-02 3.90020367e-02 ... 1.93190816e-02\n",
+ " 0.00000000e+00 4.05835544e-02]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.87213355e-01 3.54572535e-01 3.34555329e-01 ... 2.93481130e-02\n",
+ " 1.24007470e-01 5.35809019e-01]\n",
+ " [3.95939583e-01 3.50938338e-01 3.49083503e-01 ... 3.08788599e-02\n",
+ " 1.30537984e-01 5.28205128e-01]\n",
+ " [4.03950346e-01 3.40988978e-01 3.41412084e-01 ... 2.99287411e-02\n",
+ " 1.43695403e-01 5.21662246e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.69314325e-02 ... 1.94246503e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.17718941e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.76616026e-04 3.25865580e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[4.87598606e-01 3.64298481e-01 3.57942974e-01 ... 2.84243864e-02\n",
+ " 1.34628592e-01 5.05570292e-01]\n",
+ " [5.02843515e-01 3.73324397e-01 3.71181263e-01 ... 3.49168646e-02\n",
+ " 1.30381348e-01 4.95225464e-01]\n",
+ " [4.12107870e-01 3.66800715e-01 3.37270876e-01 ... 3.61045131e-02\n",
+ " 1.44478583e-01 5.13527851e-01]\n",
+ " ...\n",
+ " [0.00000000e+00 4.46827525e-04 3.32993890e-02 ... 2.09026128e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.12627291e-02 ... 1.90023753e-02\n",
+ " 0.00000000e+00 0.00000000e+00]\n",
+ " [0.00000000e+00 4.46827525e-04 3.21792261e-02 ... 2.09026128e-02\n",
+ " 0.00000000e+00 0.00000000e+00]]\n",
+ "list_RMSE_SousModele [0.07463769895332431, 0.0695089934020948, 0.06616230568613808, 0.06678827946261061, 0.070212288143011, 0.0800147124353753, 0.0733171542287429, 0.09339278103848497, 0.0727082393569627, 0.0647461161557969] mean of list_RMSE_SousModele 0.07314885688625417\n",
+ " RMSE resultat vote 0.059076848904574736\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "key 1 value[6] 0.058100387712334015\n",
+ "key 8 value[6] 0.0957523625178593\n",
+ "key 10 value[6] 0.11586412658816056\n",
+ "key 11 value[6] 0.1288294285628787\n",
+ "key 13 value[6] 0.16334850821192484\n",
+ "key 15 value[6] 0.21615527050978905\n",
+ "key 16 value[6] 0.25465259711919624\n",
+ "key 17 value[6] 0.3081780004228992\n",
+ "key 18 value[6] 0.39107838594488675\n",
+ "key 19 value[6] 0.5474158342212556\n",
+ "key 20 value[6] 1.0\n",
+ "key 1 value[2] 0.03894127863148858\n",
+ "key 8 value[2] 0.07424348233279235\n",
+ "key 10 value[2] 0.11337091103964683\n",
+ "key 11 value[2] 0.15710132087757303\n",
+ "key 13 value[2] 0.20660932107749158\n",
+ "key 15 value[2] 0.26388685939833495\n",
+ "key 16 value[2] 0.33374542525891104\n",
+ "key 17 value[2] 0.4183864493622142\n",
+ "key 18 value[2] 0.5346529546599096\n",
+ "key 19 value[2] 0.696294852068805\n",
+ "key 20 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.059642363559292144\n",
+ "Current Error err_H 0.059076848904574736\n",
+ "myFeeder.t 20\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536, 0.04963775098677459, 0.04659546922172779, 0.03983820166421492, 0.060894731847603654, 0.044176494409285595, 0.05113756750316481, 0.059076848904574736]\n",
+ "[[0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " ...\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]\n",
+ " [0. 0. 0. ... 0. 0. 0.]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.01473184 ... 0.01773555 0. 0. ]\n",
+ " [0. 0.00223414 0.0191446 ... 0.01794669 0. 0. ]\n",
+ " [0. 0.00226393 0.02043449 ... 0.01815783 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.0143924 ... 0.0178939 0. 0. ]\n",
+ " [0. 0.00044683 0.01446029 ... 0.01794669 0. 0. ]\n",
+ " [0. 0.00047662 0.01432451 ... 0.01752441 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00053619 0.02552614 ... 0.01889681 0. 0.00194518]\n",
+ " [0. 0.00563003 0.02885268 ... 0.01873845 0. 0. ]\n",
+ " [0. 0.01283884 0.03306178 ... 0.01916073 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00988978 0.02715547 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00988978 0.02715547 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00616622 0.02729124 ... 0.01821061 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03448744 ... 0.02042755 0. 0. ]\n",
+ " [0. 0.00056598 0.03394433 ... 0.02148324 0. 0. ]\n",
+ " [0. 0.0005064 0.03312967 ... 0.02032198 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.08713137 0.04236253 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.0813524 0.04202308 ... 0.01805226 0. 0. ]\n",
+ " [0. 0.05248734 0.033537 ... 0.01831618 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.0005064 0.02864902 ... 0.02306677 0. 0. ]\n",
+ " [0. 0.00044683 0.03048201 ... 0.02185273 0. 0. ]\n",
+ " [0. 0.00047662 0.034759 ... 0.02164159 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03957909 ... 0.02158881 0. 0. ]\n",
+ " [0. 0.00044683 0.03957909 ... 0.02158881 0. 0. ]\n",
+ " [0. 0.00044683 0.03957909 ... 0.02153603 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03598099 ... 0.01942465 0. 0. ]\n",
+ " [0. 0.00047662 0.03238289 ... 0.02048034 0. 0. ]\n",
+ " [0. 0.0005064 0.03374067 ... 0.01963579 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.03971487 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00044683 0.03971487 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00044683 0.03971487 ... 0.01868567 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00379803 0.03014257 ... 0.02177356 0. 0. ]\n",
+ " [0. 0.00531725 0.03014257 ... 0.01979414 0. 0. ]\n",
+ " [0. 0.00370867 0.03197556 ... 0.01908155 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00053619 0.03869654 ... 0.01908155 0. 0. ]\n",
+ " [0. 0.00058088 0.03930754 ... 0.01876485 0. 0. ]\n",
+ " [0. 0.00053619 0.03940937 ... 0.01741884 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.03204345 ... 0.01863288 0. 0. ]\n",
+ " [0. 0.00047662 0.03143245 ... 0.01910794 0. 0. ]\n",
+ " [0. 0.00044683 0.03170401 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02545825 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00044683 0.02545825 ... 0.01868567 0. 0. ]\n",
+ " [0. 0.00044683 0.02545825 ... 0.01868567 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00047662 0.0305499 ... 0.01884402 0. 0. ]\n",
+ " [0. 0.00056598 0.03095723 ... 0.01894959 0. 0. ]\n",
+ " [0. 0.00047662 0.03109301 ... 0.01921351 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00047662 0.02715547 ... 0.01894959 0. 0. ]\n",
+ " [0. 0.00047662 0.02715547 ... 0.01894959 0. 0. ]\n",
+ " [0. 0.00047662 0.02701969 ... 0.01894959 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.03156823 ... 0.0202692 0. 0. ]\n",
+ " [0. 0.00049151 0.03533605 ... 0.0202692 0. 0. ]\n",
+ " [0. 0.00053619 0.03472505 ... 0.01931908 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.02983707 ... 0.01908155 0. 0. ]\n",
+ " [0. 0.00044683 0.03105906 ... 0.01916073 0. 0. ]\n",
+ " [0. 0.00044683 0.0290224 ... 0.01987332 0. 0. ]]\n",
+ "<class 'numpy.ndarray'>\n",
+ "Output of each sub-model [[0. 0.00044683 0.04480652 ... 0.01821061 0. 0. ]\n",
+ " [0. 0.00044683 0.04175153 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.04175153 ... 0.01900238 0. 0. ]\n",
+ " ...\n",
+ " [0. 0.00044683 0.04175153 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.04175153 ... 0.01900238 0. 0. ]\n",
+ " [0. 0.00044683 0.04175153 ... 0.01900238 0. 0. ]]\n",
+ "list_RMSE_SousModele [0.0734956053950727, 0.072789049140701, 0.06811017674013405, 0.07756887207465289, 0.07510316908234548, 0.08435138708458828, 0.07774962937310508, 0.07268098949301066, 0.0517444455891126, 0.07370306533288701] mean of list_RMSE_SousModele 0.07272963893056099\n",
+ " RMSE resultat vote 0.0526199539398311\n",
+ "vote is over\n",
+ "Instance weights, recording current group performance, over\n",
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Lasso\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n",
+ "/Users/taopeng/opt/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/_coordinate_descent.py:531: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 0.0, tolerance: 0.0\n",
+ " positive)\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Start adding new sub-models\n",
+ "Randomly selected indique_sousModle Tree\n",
+ "key 1 value[6] 0.054938194722757405\n",
+ "key 8 value[6] 0.08787852233288299\n",
+ "key 10 value[6] 0.10497512198316751\n",
+ "key 11 value[6] 0.11586412658816056\n",
+ "key 13 value[6] 0.1444116730248319\n",
+ "key 15 value[6] 0.18670485439918758\n",
+ "key 16 value[6] 0.21615527050978905\n",
+ "key 18 value[6] 0.3081780004228992\n",
+ "key 19 value[6] 0.39107838594488675\n",
+ "key 20 value[6] 0.5474158342212556\n",
+ "key 21 value[6] 1.0\n",
+ "key 1 value[2] 0.03910905914860416\n",
+ "key 8 value[2] 0.07483787127152222\n",
+ "key 10 value[2] 0.11462452736476031\n",
+ "key 11 value[2] 0.15899301582294226\n",
+ "key 13 value[2] 0.20968116583615845\n",
+ "key 15 value[2] 0.26821268011384897\n",
+ "key 16 value[2] 0.33902934390336753\n",
+ "key 18 value[2] 0.4256922889807181\n",
+ "key 19 value[2] 0.5390507630200201\n",
+ "key 20 value[2] 0.6982056221574771\n",
+ "key 21 value[2] 1.0\n",
+ "Average errors so far np.mean(errList) 0.05930796310122257\n",
+ "Current Error err_H 0.0526199539398311\n",
+ "myFeeder.t 21\n",
+ "[0.09649610570283121, 0.056939832739505786, 0.07351295895389018, 0.06551391490994131, 0.09254022964979645, 0.05907000031746698, 0.06331392399873853, 0.06559129223226406, 0.06382766780448922, 0.06008641397815921, 0.05320058245529627, 0.04595840192499219, 0.04543888198112536, 0.04963775098677459, 0.04659546922172779, 0.03983820166421492, 0.060894731847603654, 0.044176494409285595, 0.05113756750316481, 0.059076848904574736, 0.0526199539398311]\n"
+ ]
+ }
+ ],
+ "source": [
+ "\n",
+ "\n",
+ "path_X_Y_data = 'simulationDonnes/'\n",
+ "Y = np.load(path_X_Y_data + 'Y_CU_BEMS_19_normale_nettoy.npy' ,allow_pickle=True)\n",
+ "X = np.load(path_X_Y_data + 'X_CU_BEMS_19_normale_nettoybackTimePoint3PCA10.npy' ,allow_pickle=True)\n",
+ "\n",
+ " \n",
+ "while(True):\n",
+ " \n",
+ " dic_expert = {} #Conserver un dictionnaire des sous-modèles\n",
+ " timeList, errList, innErrList, testErrNextBatchList,nFlodErr= [],[],[],[],[] #Liste des temps, liste de la précision \n",
+ " Resulta_list, Ture_list = [],[] #Recorded information: predicted results and true values\n",
+ " dic_expert = {} #Conserver un dictionnaire des sous-modèles\n",
+ " ini_pourCentage = 0.01 # Le rapport initial doit être défini \n",
+ " testStar,testEnd = 0.2,0.99\n",
+ " HowManyDaysForBatch =14\n",
+ " batch_size = 1400*HowManyDaysForBatch# Taille du bloc\n",
+ " a,b = 0.35,1\n",
+ " time_Ini = 0 # Le temps nécessaire à l'initialisation est enregistré ici\n",
+ " varepsilon_extreme_bord = 1.2 # Coefficient permettant de vérifier si le sous-modèle fonctionne correctement\n",
+ " maxNumSousModle = 10 # Nombre maximal de sous-modèles\n",
+ " testBatchNum = 58 #Contrôler la quantité de circulation pendant l'expérience\n",
+ " Q3 = 75 # Définir les coefficients pour le calcul des poids d'instance\n",
+ " optionPCA, PCA_size = False, 25\n",
+ "# optitionInnErrOrCroissErr = 'InnErr'\n",
+ " optitionInnErrOrCroissErr = 'CroissErr'\n",
+ " optitionAddSousModele = True\n",
+ " optionVote ='Mean'\n",
+ " # optionVote ='OnlyMaxERRWeight'\n",
+ " # optionVote = 'vote_OnlyMaxESpWeight'\n",
+ " indicateur_sousModel = 'Random_LI_R_T_KNN' #['Random','LSTM','Lasso','KNN','R_Forest'] # 0 随机，1 LSTM，\n",
+ " cStepAugment = 1.4\n",
+ " updateOrNon = False\n",
+ " \n",
+ " myFeeder = feeder_Ini_Train_Batch(X,Y,testStar,testEnd,batch_size) # X,Y beginCentage,endCentage,batch_size \n",
+ " X_train,Y_train = myFeeder.getIni_X_Y(ini_pourCentage) #这里才设定初始比例\n",
+ " X_test,Y_test = myFeeder.getTrain_X_Y()\n",
+ "\n",
+ " iniFlage = True\n",
+ " while ((testBatchNum > myFeeder.t) and myFeeder.hasThisBatch()):\n",
+ "\n",
+ " # Créer un sous-modèle initialisé, code 0\n",
+ " start_Ini = time.process_time()# Heure de début d'enregistrement记录开始时间\n",
+ " if iniFlage == True:\n",
+ " Delta = [1/ myFeeder.batch_size]* myFeeder.batch_size #Valeur par défaut de delta delta的默认值 \n",
+ " dic_expert[0], err_numFlod= RandonSelectionModle(0,X_train, Y_train,optitionInnErrOrCroissErr, indique_sousModle = indicateur_sousModel)\n",
+ " iniFlage = False\n",
+ " time_Ini = round(time.process_time()-start_Ini,5) # Le temps nécessaire à l'initialisation est enregistré ici 这里记录了初始化需要的时间\n",
+ " start = time.process_time()# Heure de début d'enregistrement 记录开始时\n",
+ " \n",
+ " #----------------VOTE------------------------------------------------------\n",
+ " actul_Batch_X, actul_Batch_Y = myFeeder.getThisBatch()\n",
+ " # Calculer la performance de Ht-1 à t \n",
+ " if optionVote == 'Mean':\n",
+ " yhat_H = vote_mean(dic_expert,actul_Batch_X, actul_Batch_Y,afficher_detail=True)\n",
+ " elif optionVote == 'Max':\n",
+ " yhat_H = vote_OnlyMaxWeight(dic_expert,actul_Batch_X, actul_Batch_Y)\n",
+ "\n",
+ " print('vote is over')\n",
+ " #----------------------------------------------------------------------\n",
+ "\n",
+ " #----------------Confirm instance weights and record current cluster performance------------------------------------------------------\n",
+ " ERR_absolu_H = np.abs(actul_Batch_Y - yhat_H) #ERR_absolu (martix)\n",
+ " Resulta_list.append(yhat_H)\n",
+ " Ture_list.append(actul_Batch_Y)\n",
+ " err_H = aRMSE(actul_Batch_Y,yhat_H)# err as RMSE \n",
+ " meanERR_absolu_H = np.mean(ERR_absolu_H, axis=1)\n",
+ " # Delta = getMeanSuperPourCentage_Martix_ParLine(ERR_absolu_H,Q3)\n",
+ " # varepsilon_H = np.average(meanERR_absolu_H,weights=Delta)\n",
+ "\n",
+ " varepsilon_H = np.average(meanERR_absolu_H)\n",
+ " errList.append(err_H) \n",
+ " print('Instance weights, recording current group performance, over')\n",
+ " #----------------确认实例权重，记录当前群性能------------------------------------------------------\n",
+ "\n",
+ " #----------------Alarm in case of abnormality------------------------------------------------------\n",
+ " if err_H >10:\n",
+ " print('time: ',myFeeder.t,'Une erreur majeure s est produite：',err_H) \n",
+ " #----------------当群性能异常时，报警------------------------------------------------------\n",
+ "\n",
+ "\n",
+ " #------------------Add a new sub-model and confirm its weights (from cross-validation)-----------------------------\n",
+ " # 添加一个新的子模型 \n",
+ " # 这里有新改动，未在 meanERR_absolu < 0.1\n",
+ " bord = 1 \n",
+ " while(optitionAddSousModele):\n",
+ " print('Start adding new sub-models')\n",
+ " dic_expert[myFeeder.t], yhat_new_numFlod = RandonSelectionModle(myFeeder.t, actul_Batch_X, actul_Batch_Y, optitionInnErrOrCroissErr,indique_sousModle = indicateur_sousModel)\n",
+ "\n",
+ " \n",
+ " if optitionInnErrOrCroissErr == 'CroissErr':\n",
+ " ERR_absolu_new_numFlod = np.abs(actul_Batch_Y - yhat_new_numFlod) \n",
+ " meanERR_absolu_new_numFlod = np.mean(ERR_absolu_new_numFlod, axis=1)\n",
+ " err_new_numFlod = aRMSE(actul_Batch_Y,yhat_new_numFlod) \n",
+ " varepsilon_new = np.average(meanERR_absolu_new_numFlod,weights=Delta)\n",
+ " if varepsilon_new < varepsilon_H*bord:\n",
+ " nFlodErr.append(err_new_numFlod)\n",
+ " break\n",
+ " else:\n",
+ " bord = bord*cStepAugment\n",
+ " \n",
+ " elif optitionInnErrOrCroissErr == 'InnErr':\n",
+ "\n",
+ " yhat_new_inner = makePredictionModele(dic_expert[myFeeder.t][0], actul_Batch_X)\n",
+ " ERR_absolu_new_inner = np.abs(actul_Batch_Y - yhat_new_inner) \n",
+ " meanERR_absolu_new_inner = np.mean(ERR_absolu_new_inner, axis=1)\n",
+ " err_new_inner = aRMSE(actul_Batch_Y,yhat_new_inner) \n",
+ " varepsilon_new = np.average(meanERR_absolu_new_inner,weights=Delta)\n",
+ " break\n",
+ " #------------------添加一个新的子模型 ，确认其权重（来自交叉验证）-----------------------------\n",
+ " \n",
+ " if optitionAddSousModele:\n",
+ " #------------------Recording training errors for new sub-models-----------------------------\n",
+ " yhat_new_inner = makePredictionModele(dic_expert[myFeeder.t][0], actul_Batch_X)\n",
+ " err_new_inner = aRMSE(actul_Batch_Y,yhat_new_inner) \n",
+ " innErrList.append(err_new_inner)\n",
+ " #------------------对新的子模型记录训练错误-----------------------------\n",
+ "\n",
+ " #------------------Test error of the model generated by the previous block in the next block-----------------------------\n",
+ " if myFeeder.hasThisBatch_and_nextBath():\n",
+ " next_batch_test_X, next_batch_test_Y = myFeeder.getNextBatch_getThisBatch()\n",
+ " yhat_next_batch_test = makePredictionModele(dic_expert[myFeeder.t][0], next_batch_test_X)\n",
+ " err_test_NextBatch = aRMSE(next_batch_test_Y, yhat_next_batch_test) \n",
+ " testErrNextBatchList.append(err_test_NextBatch)\n",
+ " #------------------对上一个块的子模型在下一块的测试错误-----------------------------\n",
+ "\n",
+ "\n",
+ " #------------------Calculate the performance of the old sub-model based on the weighted strengths in the current block-----------\n",
+ " Varepsilon_list = [] #Collecte de la liste des pesées séparées de Varepsilon pour faciliter la recherche des valeurs max-min. 收集Varepsilon单独称list，方便找最大最小值\n",
+ " for key,value in dic_expert.items():\n",
+ " yhat_oldModel = makePredictionModele(value[0], actul_Batch_X)\n",
+ " ERR_absolu_oldModel = np.abs(actul_Batch_Y - yhat_oldModel)\n",
+ " mean_ERR_absolu_oldModel = np.mean(ERR_absolu_oldModel, axis=1)\n",
+ " err_oldModel = aRMSE(actul_Batch_Y,yhat_oldModel) # RMSE\n",
+ " varepsilon_oldModel = np.average(mean_ERR_absolu_oldModel,weights=Delta)\n",
+ " # if varepsilon_oldModel > varepsilon_H*varepsilon_extreme_bord: \n",
+ " # value[5] = varepsilon_H*varepsilon_extreme_bord\n",
+ " # else:\n",
+ " # value[5] = varepsilon_oldModel\n",
+ " value[5] = varepsilon_oldModel\n",
+ " Varepsilon_list.append(value[5])\n",
+ " value[7] = err_oldModel # enregrister RMSE\n",
+ " # for key,value in dic_expert.items(): #normalisation \n",
+ " # value[5] = (value[5] - min(Varepsilon_list))/(max(Varepsilon_list) - min(Varepsilon_list))\n",
+ " # for key,value in dic_expert.items(): \n",
+ " # print('key ', key, ' value[5] ', value[5] )\n",
+ " #------------------根据当前块中被加权的实力，计算旧的子模型的表现（7号普通错误），并更新权重（5号根据加权错误）-----------\n",
+ "\n",
+ "\n",
+ " #------------------Calculating time weights ----------\n",
+ " for key,value in dic_expert.items(): \n",
+ " numerator = beta_fonction(value[1],myFeeder.t,a,b)\n",
+ " denominator = 0\n",
+ " for j in range(0, myFeeder.t - value[1] +1): # \n",
+ " denominator += beta_fonction(myFeeder.t - j,myFeeder.t,a,b)\n",
+ "# print('denominator ' , denominator)\n",
+ "# if denominator == 0:\n",
+ "# print(' myFeeder.t ', myFeeder.t,' value[1]', value[1] )\n",
+ "# for j in range(0, myFeeder.t - value[1] +1): # \n",
+ "# print('j :', j , ' beta_fonction(myFeeder.t - j,myFeeder.t,a,b)', beta_fonction(myFeeder.t - j,myFeeder.t,a,b))\n",
+ " value[6] = numerator/denominator\n",
+ " for key,value in dic_expert.items(): \n",
+ " print('key ', key, ' value[6] ', value[6] )\n",
+ " #------------------计算6 Omega, 也就是时间权重 omega越小，越重要-----------\n",
+ "\n",
+ "\n",
+ " #------------------Calculating sub-model weights-----------\n",
+ " # Calculer les poids des sous-modèles\n",
+ " weight_list = []\n",
+ " for key,value in dic_expert.items(): \n",
+ " denominator = 0 \n",
+ " for j in range(0, myFeeder.t - value[1] +1): # from 0 to t-k\n",
+ " if ((myFeeder.t-j) in dic_expert.keys()):\n",
+ " denominator += dic_expert[myFeeder.t-j][6]*dic_expert[myFeeder.t-j][5]+0.1 # Omega 权重 * 5 当前块被加权的错误 \n",
+ " value[2] = math.log(1/denominator) \n",
+ " weight_list.append(math.log(1/denominator)) \n",
+ " for key,value in dic_expert.items(): \n",
+ " value[2] = (value[2] - min(weight_list)+0.1)/(max(weight_list) - min(weight_list)+0.1)\n",
+ " for key,value in dic_expert.items(): \n",
+ " print('key ', key, 'value[2] ',value[2] )\n",
+ " #------------------Calculating sub-model weights-----------\n",
+ "\n",
+ "\n",
+ " #------------------Removing the worst model -----------\n",
+ " if len(dic_expert) > maxNumSousModle:\n",
+ " maxErr = 0\n",
+ " del_key = 0\n",
+ " for key,value in dic_expert.items(): \n",
+ " if dic_expert[key][7] > maxErr:\n",
+ " maxErr = dic_expert[key][7]\n",
+ " del_key = key\n",
+ " dic_expert.pop(del_key)\n",
+ " #------------------去掉模型 最差策略 Sous-modèle de suppression de la pire stratégie-----------\n",
+ " \n",
+ " \n",
+ " #------------------Whether the sub-model LSTM is updated-----------\n",
+ " if updateOrNon:\n",
+ " updatingALLSousModele(dic_expert,actul_Batch_X,actul_Batch_Y) \n",
+ " #------------------子模型LSTM是否更新-----------\n",
+ "\n",
+ "\n",
+ " # 记录时间 Durée d'enregistrement\n",
+ " timeList.append(round(time.process_time()-start,3)) #记录时间\n",
+ "\n",
+ " print('Average errors so far np.mean(errList)', np.mean(errList)) #展示平均\n",
+ " print('Current Error err_H', err_H)\n",
+ "\n",
+ " #进入下一块\n",
+ " print('myFeeder.t ', myFeeder.t)\n",
+ " myFeeder.goNext()\n",
+ "\n",
+ " print(errList)\n",
+ "\n",
+ " break;\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.7.6"
+ },
+ "varInspector": {
+ "cols": {
+ "lenName": 16,
+ "lenType": 16,
+ "lenVar": 40
+ },
+ "kernels_config": {
+ "python": {
+ "delete_cmd_postfix": "",
+ "delete_cmd_prefix": "del ",
+ "library": "var_list.py",
+ "varRefreshCmd": "print(var_dic_list())"
+ },
+ "r": {
+ "delete_cmd_postfix": ") ",
+ "delete_cmd_prefix": "rm(",
+ "library": "var_list.r",
+ "varRefreshCmd": "cat(var_dic_list()) "
+ }
+ },
+ "position": {
+ "height": "662px",
+ "left": "601px",
+ "right": "20px",
+ "top": "121px",
+ "width": "800px"
+ },
+ "types_to_exclude": [
+ "module",
+ "function",
+ "builtin_function_or_method",
+ "instance",
+ "_Feature"
+ ],
+ "window_display": false
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/__pycache__/msvr.cpython-37.pyc b/__pycache__/msvr.cpython-37.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..abf8cfccd50a1022842fbc521c3e3ece984da871
Binary files /dev/null and b/__pycache__/msvr.cpython-37.pyc differ
diff --git a/msvr.py b/msvr.py
new file mode 100644
index 0000000000000000000000000000000000000000..91ac99ab6ad524cba334ea679994fe84625d12da
--- /dev/null
+++ b/msvr.py
@@ -0,0 +1,198 @@
+#!/usr/bin/env python
+# coding: utf-8
+import numpy as np
+
+'''
+Inputs:
+ x : training patterns (num_samples * n_d),
+ y : training targets (num_samples * n_k),
+ ker : kernel type ('lin', 'poly', 'rbf'),
+ C : cost parameter,
+ par : kernel parameter (see function 'kernelmatrix'),
+ tol : tolerance.
+Outputs:
+ Beta
+'''
+
+def msvr(x, y, ker, C, epsi, par, tol):
+ n_m = np.shape(x)[0] #num of samples
+ n_d = np.shape(x)[1] #input data dimensionality
+ n_k = np.shape(y)[1] #output data dimensionality (output variables)
+
+ #build the kernel matrix on the labeled samples
+ H = kernelmatrix(ker, x, x, par)
+
+ #create martix for regression parameters
+ Beta = np.zeros((n_m, n_k))
+
+ #E = prediction error per output (n_m * n_k)
+ E = y - np.dot(H, Beta)
+ #RSE
+ u = np.sqrt(np.sum(E**2,1,keepdims=True))
+
+ #RMSE
+ RMSE = []
+ RMSE_0 = np.sqrt(np.mean(u**2))
+ RMSE.append(RMSE_0)
+
+ #points for which prediction error is larger than epsilon
+ i1 = np.where(u>epsi)[0]
+
+ #set initial values of alphas a (n_m * 1)
+ a = 2 * C * (u - epsi) / u
+
+ #L (n_m * 1)
+ L = np.zeros(u.shape)
+
+ # we modify only entries for which u > epsi. with the sq slack
+ L[i1] = u[i1]**2 - 2 * epsi * u[i1] + epsi**2
+
+ #Lp is the quantity to minimize (sq norm of parameters + slacks)
+ Lp = []
+ BetaH = np.dot(np.dot(Beta.T, H), Beta)
+ Lp_0 = np.sum(np.diag(BetaH), 0) / 2 + C * np.sum(L)/2
+ Lp.append(Lp_0)
+
+ eta = 1
+ k = 1
+ hacer = 1
+ val = 1
+
+ while(hacer):
+ Beta_a = Beta.copy()
+ E_a = E.copy()
+ u_a = u.copy()
+ i1_a = i1.copy()
+
+ M1 = H[i1][:,i1] + np.diagflat(1/a[i1]) + 1e-10 * np.eye(len(a[i1]))
+
+ #compute betas
+# sal1 = np.dot(np.linalg.pinv(M1),y[i1]) #求逆or广义逆（M-P逆）无法保证M1一定是可逆的？
+ sal1 = np.dot(np.linalg.inv(M1),y[i1])
+
+ eta = 1
+ Beta = np.zeros(Beta.shape)
+ Beta[i1] = sal1.copy()
+
+ #error
+ E = y - np.dot(H, Beta)
+ #RSE
+ u = np.sqrt(np.sum(E**2,1)).reshape(n_m,1)
+ i1 = np.where(u>=epsi)[0]
+
+ L = np.zeros(u.shape)
+ L[i1] = u[i1]**2 - 2 * epsi * u[i1] + epsi**2
+
+ #%recompute the loss function
+ BetaH = np.dot(np.dot(Beta.T, H), Beta)
+ Lp_k = np.sum(np.diag(BetaH), 0) / 2 + C * np.sum(L)/2
+ Lp.append(Lp_k)
+
+ #Loop where we keep alphas and modify betas
+ while(Lp[k] > Lp[k-1]):
+ eta = eta/10
+ i1 = i1_a.copy()
+
+ Beta = np.zeros(Beta.shape)
+ #%the new betas are a combination of the current (sal1)
+ #and of the previous iteration (Beta_a)
+ Beta[i1] = eta*sal1 + (1-eta)*Beta_a[i1]
+
+ E = y - np.dot(H, Beta)
+ u = np.sqrt(np.sum(E**2,1)).reshape(n_m,1)
+
+ i1 = np.where(u>=epsi)[0]
+
+ L = np.zeros(u.shape)
+ L[i1] = u[i1]**2 - 2 * epsi * u[i1] + epsi**2
+ BetaH = np.dot(np.dot(Beta.T, H), Beta)
+ Lp_k = np.sum(np.diag(BetaH), 0) / 2 + C * np.sum(L)/2
+ Lp[k] = Lp_k
+
+ #stopping criterion 1
+ if(eta < 1e-16):
+ Lp[k] = Lp[k-1]- 1e-15
+ Beta = Beta_a.copy()
+
+ u = u_a.copy()
+ i1 = i1_a.copy()
+
+ hacer = 0
+
+ #here we modify the alphas and keep betas
+ a_a = a.copy()
+ a = 2 * C * (u - epsi) / u
+
+ RMSE_k = np.sqrt(np.mean(u**2))
+ RMSE.append(RMSE_k)
+
+ if((Lp[k-1]-Lp[k])/Lp[k-1] < tol):
+ hacer = 0
+
+ k = k + 1
+
+ #stopping criterion #algorithm does not converge. (val = -1)
+ if(len(i1) == 0):
+ hacer = 0
+ Beta = np.zeros(Beta.shape)
+ val = -1
+
+ NSV = len(i1)
+
+ return Beta
+
+'''
+KERNELMATRIX
+
+Builds a kernel from training and test data matrices.
+
+Inputs:
+ ker: {'lin' 'poly' 'rbf'}
+ X: Xtest (num_test * n_d)
+ X2: Xtrain (num_train * n_d)
+ parameter:
+ width of the RBF kernel
+ bias in the linear and polinomial kernel
+ degree in the polynomial kernel
+
+Output:
+ K: kernel matrix
+'''
+
+def kernelmatrix(ker, X, X2, p=0):
+
+ X = X.T
+ X2 = X2.T
+
+ if(ker == 'lin'):
+ tmp1, XX2_norm, tmp2 = np.linalg.svd(np.dot(X.T,X2))
+ XX2_norm = np.max(XX2_norm)
+ K = np.dot(X.T,X2)/XX2_norm
+
+ elif(ker == 'poly'):
+ tmp1, XX2_norm, tmp2 = np.linalg.svd(np.dot(X.T,X2))
+ XX2_norm = np.max(XX2_norm)
+ K = (np.dot(X.T,X2)/XX2_norm*p[0] + p[1]) ** p[2]
+
+ elif(ker == 'rbf'):
+ n1sq = np.sum(X**2,0,keepdims=True)
+ n1 = X.shape[1]
+
+ if(n1 == 1): #just one feature
+ N1 = X.shape[0]
+ N2 = X2.shape[0]
+ D = np.zeros((N1,N2))
+ for i in range(0,N1):
+ D[i] = (X2 - np.dot(np.ones((N2,1)),X[i].reshape(1,-1))).T * (X2 - np.dot(np.ones((N2,1)),X[i].reshape(1,-1))).T
+ else:
+ n2sq = np.sum(X2**2,0,keepdims=True)
+ n2 = X2.shape[1]
+ D = (np.dot(np.ones((n2,1)),n1sq)).T + np.dot(np.ones((n1,1)),n2sq) - 2*np.dot(X.T, X2)
+
+ K = np.exp((-D**2)/(2*p**2))
+
+ else:
+ print("no such kernel")
+ K = 0
+
+ return K
diff --git a/simulationDonnes/X_CU_BEMS_19_normale_nettoybackTimePoint3PCA10.npy b/simulationDonnes/X_CU_BEMS_19_normale_nettoybackTimePoint3PCA10.npy
new file mode 100644
index 0000000000000000000000000000000000000000..7da89d8088f9909748232b0742434df47ac41e0b
Binary files /dev/null and b/simulationDonnes/X_CU_BEMS_19_normale_nettoybackTimePoint3PCA10.npy differ
diff --git a/simulationDonnes/Y_CU_BEMS_19_normale_nettoy.npy b/simulationDonnes/Y_CU_BEMS_19_normale_nettoy.npy
new file mode 100644
index 0000000000000000000000000000000000000000..797c0c581fccaf470c3dcbc469609a2fbad8f58e
Binary files /dev/null and b/simulationDonnes/Y_CU_BEMS_19_normale_nettoy.npy differ