From fff1e74e552cc87ddbe453fb22385a0251863ece Mon Sep 17 00:00:00 2001
From: bbauvin <baptiste.bauvin@centrale-marseille.fr>
Date: Thu, 5 Oct 2017 11:13:13 -0400
Subject: [PATCH] Tested randomstate on monoview replaced by scipy.stats
distribs seems to be reproductible
---
.../Monoview/ExecClassifMonoView.py | 2 +-
.../MonoviewClassifiers/Adaboost.py | 6 +++---
.../MonoviewClassifiers/DecisionTree.py | 4 ++--
.../MonoviewClassifiers/KNN.py | 3 ++-
.../MonoviewClassifiers/RandomForest.py | 6 +++---
.../MonoviewClassifiers/SCM.py | 10 ++++++----
.../MonoviewClassifiers/SGD.py | 4 +++-
.../MonoviewClassifiers/SVMLinear.py | 3 ++-
.../MonoviewClassifiers/SVMPoly.py | 6 ++++--
.../MonoviewClassifiers/SVMRBF.py | 3 ++-
.../Multiview/ExecMultiview.py | 4 ++--
.../Multiview/Fusion/Fusion.py | 2 +-
.../Multiview/GetMultiviewDb.py | 2 +-
Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py | 2 +-
14 files changed, 33 insertions(+), 24 deletions(-)
diff --git a/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py b/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
index 58445abf..9a162b00 100644
--- a/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
+++ b/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
@@ -21,7 +21,7 @@ import ExportResults # Functions to render results
import MonoviewClassifiers
import Metrics
from analyzeResult import execute
-from ..utils.Dataset import getV, getValue, extractSubset
+from utils.Dataset import getV, getValue, extractSubset
# Author-Info
__author__ = "Nikolas Huelsmann, Baptiste BAUVIN"
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
index 567cfa78..54fa6cd2 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
@@ -3,6 +3,7 @@ from sklearn.pipeline import Pipeline
from sklearn.model_selection import RandomizedSearchCV
from sklearn.tree import DecisionTreeClassifier
import Metrics
+from scipy.stats import randint
# Author-Info
__author__ = "Baptiste Bauvin"
@@ -16,8 +17,7 @@ def canProbas():
def fit(DATASET, CLASS_LABELS, randomState, NB_CORES=1,**kwargs):
num_estimators = int(kwargs['0'])
base_estimators = DecisionTreeClassifier()#kwargs['1']
- classifier = AdaBoostClassifier(n_estimators=num_estimators, base_estimator=base_estimators,
- random_state=randomState)
+ classifier = AdaBoostClassifier(n_estimators=num_estimators, base_estimator=base_estimators, random_state=randomState)
classifier.fit(DATASET, CLASS_LABELS)
return classifier
@@ -35,7 +35,7 @@ def getKWARGS(kwargsList):
def randomizedSearch(X_train, y_train, randomState, nbFolds=4, metric=["accuracy_score", None], nIter=30, nbCores=1):
pipeline = Pipeline([('classifier', AdaBoostClassifier())])
- param= {"classifier__n_estimators": randomState.randint(1, 15),
+ param= {"classifier__n_estimators": randint(1, 15),
"classifier__base_estimator": [DecisionTreeClassifier()]}
metricModule = getattr(Metrics, metric[0])
if metric[1]!=None:
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
index 2c9cbc4b..a378bb0a 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
@@ -2,7 +2,7 @@ from sklearn.tree import DecisionTreeClassifier
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.model_selection import RandomizedSearchCV
import Metrics
-
+from scipy.stats import randint
# Author-Info
__author__ = "Baptiste Bauvin"
@@ -37,7 +37,7 @@ def getKWARGS(kwargsList):
def randomizedSearch(X_train, y_train, randomState, nbFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_DT = Pipeline([('classifier', DecisionTreeClassifier())])
- param_DT = {"classifier__max_depth": randomState.randint(1, 30),
+ param_DT = {"classifier__max_depth": randint(1, 30),
"classifier__criterion": ["gini", "entropy"],
"classifier__splitter": ["best", "random"]}
metricModule = getattr(Metrics, metric[0])
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py
index 8953721c..fd16f1fb 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py
@@ -2,6 +2,7 @@ from sklearn.neighbors import KNeighborsClassifier
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.model_selection import RandomizedSearchCV
import Metrics
+from scipy.stats import randint
# Author-Info
@@ -40,7 +41,7 @@ def getKWARGS(kwargsList):
def randomizedSearch(X_train, y_train, randomState, nbFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30 ):
pipeline_KNN = Pipeline([('classifier', KNeighborsClassifier())])
- param_KNN = {"classifier__n_neighbors": randomState.randint(1, 50),
+ param_KNN = {"classifier__n_neighbors": randint(1, 50),
"classifier__weights": ["uniform", "distance"],
"classifier__algorithm": ["auto", "ball_tree", "kd_tree", "brute"],
"classifier__p": [1,2],
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
index ee85e6fd..fe95c8da 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
@@ -2,7 +2,7 @@ from sklearn.ensemble import RandomForestClassifier
from sklearn.pipeline import Pipeline
from sklearn.model_selection import RandomizedSearchCV
import Metrics
-
+from scipy.stats import randint
# Author-Info
__author__ = "Baptiste Bauvin"
@@ -37,8 +37,8 @@ def getKWARGS(kwargsList):
def randomizedSearch(X_train, y_train, randomState, nbFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_rf = Pipeline([('classifier', RandomForestClassifier())])
- param_rf = {"classifier__n_estimators": randomState.randint(1, 30),
- "classifier__max_depth": randomState.randint(1, 30),
+ param_rf = {"classifier__n_estimators": randint(1, 30),
+ "classifier__max_depth": randint(1, 30),
"classifier__criterion": ["gini", "entropy"]}
metricModule = getattr(Metrics, metric[0])
if metric[1]!=None:
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SCM.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SCM.py
index afc365b9..06668c7e 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SCM.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SCM.py
@@ -1,11 +1,13 @@
import Metrics
from pyscm.utils import _pack_binary_bytes_to_ints
import pyscm
-from ..utils.Dataset import getShape
+from utils.Dataset import getShape
import h5py
from Multiview import GetMultiviewDb as DB
from pyscm.binary_attributes.base import BaseBinaryAttributeList
import os
+
+
# Author-Info
__author__ = "Baptiste Bauvin"
__status__ = "Prototype" # Production, Development, Prototype
@@ -15,7 +17,7 @@ __status__ = "Prototype" # Production, Development, P
def canProbas():
return False
-def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
+def fit(DATASET, CLASS_LABELS, randomState, NB_CORES=1,**kwargs):
max_attrtibutes = kwargs['0']
try:
p = kwargs['1']
@@ -68,11 +70,11 @@ def randomizedSearch(X_train, y_train, randomState, nbFolds=4, metric=["accuracy
config = []
for iterIndex in range(nIter):
max_attributes = randomState.randint(1, 20)
- p = randomState.random()
+ p = randomState.random_sample()
model = randomState.choice(["conjunction", "disjunction"])
classifier = pyscm.scm.SetCoveringMachine(p=p, max_attributes=max_attributes, model_type=model, verbose=False)
if nbFolds != 1:
- kFolds = DB.getKFoldIndices(nbFolds, y_train, len(set(y_train)), range(len(y_train)))
+ kFolds = DB.getKFoldIndices(nbFolds, y_train, len(set(y_train)), range(len(y_train)), randomState)
else:
kFolds = [[], range(len(y_train))]
scores = []
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
index 3d861549..4006470a 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
@@ -2,6 +2,8 @@ from sklearn.linear_model import SGDClassifier
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.model_selection import RandomizedSearchCV
import Metrics
+from scipy.stats import uniform
+
# Author-Info
@@ -41,7 +43,7 @@ def randomizedSearch(X_train, y_train, randomState, nbFolds=4, nbCores=1, metric
pipeline_SGD = Pipeline([('classifier', SGDClassifier())])
losses = ['log', 'modified_huber']
penalties = ["l1", "l2", "elasticnet"]
- alphas = randomState.uniform()
+ alphas = uniform()
param_SGD = {"classifier__loss": losses, "classifier__penalty": penalties,
"classifier__alpha": alphas}
metricModule = getattr(Metrics, metric[0])
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py
index 4dab4815..2e378edd 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py
@@ -2,6 +2,7 @@ from sklearn.svm import SVC
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.model_selection import RandomizedSearchCV
import Metrics
+from scipy.stats import randint
# Author-Info
@@ -30,7 +31,7 @@ def getKWARGS(kwargsList):
def randomizedSearch(X_train, y_train, randomState, nbFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_SVMLinear = Pipeline([('classifier', SVC(kernel="linear", max_iter=1000))])
- param_SVMLinear = {"classifier__C": randomState.randint(1, 10000)}
+ param_SVMLinear = {"classifier__C": randint(1, 10000)}
metricModule = getattr(Metrics, metric[0])
if metric[1]!=None:
metricKWARGS = dict((index, metricConfig) for index, metricConfig in enumerate(metric[1]))
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
index 454ad828..7b724173 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
@@ -2,6 +2,8 @@ from sklearn.svm import SVC
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.model_selection import RandomizedSearchCV
import Metrics
+from scipy.stats import randint
+
# Author-Info
@@ -33,8 +35,8 @@ def getKWARGS(kwargsList):
def randomizedSearch(X_train, y_train, randomState, nbFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_SVMPoly = Pipeline([('classifier', SVC(kernel="poly", max_iter=1000))])
- param_SVMPoly = {"classifier__C": randomState.randint(1, 10000),
- "classifier__degree": randomState.randint(1, 30)}
+ param_SVMPoly = {"classifier__C": randint(1, 10000),
+ "classifier__degree": randint(1, 30)}
metricModule = getattr(Metrics, metric[0])
if metric[1]!=None:
metricKWARGS = dict((index, metricConfig) for index, metricConfig in enumerate(metric[1]))
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py
index 24cc8681..7e32e9cc 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py
@@ -2,6 +2,7 @@ from sklearn.svm import SVC
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.model_selection import RandomizedSearchCV
import Metrics
+from scipy.stats import randint
# Author-Info
@@ -30,7 +31,7 @@ def getKWARGS(kwargsList):
def randomizedSearch(X_train, y_train, randomState, nbFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_SVMRBF = Pipeline([('classifier', SVC(kernel="rbf", max_iter=1000))])
- param_SVMRBF = {"classifier__C": randomState.randint(1, 10000)}
+ param_SVMRBF = {"classifier__C": randint(1, 10000)}
metricModule = getattr(Metrics, metric[0])
if metric[1]!=None:
metricKWARGS = dict((index, metricConfig) for index, metricConfig in enumerate(metric[1]))
diff --git a/Code/MonoMutliViewClassifiers/Multiview/ExecMultiview.py b/Code/MonoMutliViewClassifiers/Multiview/ExecMultiview.py
index 153b0b6a..e5f723a8 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/ExecMultiview.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/ExecMultiview.py
@@ -11,8 +11,8 @@ import os
import logging
import time
import h5py
-from ..utils.Dataset import getShape
-from ..utils.HyperParameterSearch import searchBestSettings
+from utils.Dataset import getShape
+from utils.HyperParameterSearch import searchBestSettings
# Author-Info
__author__ = "Baptiste Bauvin"
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
index 11a58af6..8e6349d3 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
@@ -66,7 +66,7 @@ def getBenchmark(benchmark, args=None):
return benchmark
-def getArgs(args, benchmark, views, viewsIndices):
+def getArgs(args, benchmark, views, viewsIndices, randomState):
if not "Monoview" in benchmark and not args.FU_L_select_monoview in ["randomClf", "Determined"]:
args.FU_L_select_monoview = "randomClf"
argumentsList = []
diff --git a/Code/MonoMutliViewClassifiers/Multiview/GetMultiviewDb.py b/Code/MonoMutliViewClassifiers/Multiview/GetMultiviewDb.py
index d3b04b43..db98411b 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/GetMultiviewDb.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/GetMultiviewDb.py
@@ -155,7 +155,7 @@ def getKFoldIndices(nbFolds, CLASS_LABELS, NB_CLASS, learningIndices, randomStat
for foldIndex, fold in enumerate(nbTrainingExamples):
trainingExamplesIndices.append([])
while fold != [0 for i in range(NB_CLASS)]:
- index = randomState.randint(0, len(learningIndices) - 1)
+ index = randomState.randint(0, len(learningIndices))
if learningIndices[index] not in usedIndices:
isUseFull, fold = isUseful(fold, learningIndices[index], CLASS_LABELS, labelDict)
if isUseFull:
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
index 6e62ad34..7709a558 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
@@ -31,7 +31,7 @@ def getBenchmark(benchmark, args=None):
return benchmark
-def getArgs(args, benchmark, views, viewsIndices):
+def getArgs(args, benchmark, views, viewsIndices, randomState):
argumentsList = []
arguments = {"CL_type": "Mumbo",
--
GitLab