From 3305fc0a87b2a68f44f4e88792670576384bc1cf Mon Sep 17 00:00:00 2001
From: bbauvin <baptiste.bauvin@centrale-marseille.fr>
Date: Thu, 12 Oct 2017 14:17:03 -0400
Subject: [PATCH] Added Heatmaps for monoview and corrected getClassifiers
methods for lateFusion
---
Code/MonoMutliViewClassifiers/ExecClassif.py | 46 +++++--
.../Monoview/ExecClassifMonoView.py | 31 +++--
.../MonoviewClassifiers/Adaboost.py | 37 +++++-
.../MonoviewClassifiers/DecisionTree.py | 11 +-
.../MonoviewClassifiers/KNN.py | 13 +-
.../MonoviewClassifiers/RandomForest.py | 11 +-
.../MonoviewClassifiers/SCM.py | 15 ++-
.../MonoviewClassifiers/SGD.py | 12 +-
.../MonoviewClassifiers/SVMLinear.py | 11 +-
.../MonoviewClassifiers/SVMPoly.py | 15 ++-
.../MonoviewClassifiers/SVMRBF.py | 11 +-
.../Multiview/Fusion/Fusion.py | 4 +-
.../EarlyFusionPackage/WeightedLinear.py | 2 +-
.../Multiview/Fusion/Methods/LateFusion.py | 125 +++++++++++-------
.../LateFusionPackage/BayesianInference.py | 4 +-
.../LateFusionPackage/MajorityVoting.py | 4 +-
.../Methods/LateFusionPackage/SCMForLinear.py | 4 +-
.../Methods/LateFusionPackage/SVMForLinear.py | 4 +-
.../LateFusionPackage/WeightedLinear.py | 4 +-
.../Multiview/Mumbo/Mumbo.py | 2 +-
.../ResultAnalysis.py | 3 +-
.../utils/HyperParameterSearch.py | 35 +++++
22 files changed, 300 insertions(+), 104 deletions(-)
diff --git a/Code/MonoMutliViewClassifiers/ExecClassif.py b/Code/MonoMutliViewClassifiers/ExecClassif.py
index 4014ce75..6b5d3781 100644
--- a/Code/MonoMutliViewClassifiers/ExecClassif.py
+++ b/Code/MonoMutliViewClassifiers/ExecClassif.py
@@ -217,13 +217,13 @@ def lateFusionSetArgs(views, viewsIndices, classes, method,
return arguments
-def initMultiviewArguments(args, benchmark, views, viewsIndices, argumentDictionaries, randomState, directory, resultsMonoview):
+def initMultiviewArguments(args, benchmark, views, viewsIndices, argumentDictionaries, randomState, directory, resultsMonoview, classificationIndices):
multiviewArguments = []
if "Multiview" in benchmark:
for multiviewAlgoName in benchmark["Multiview"]:
multiviewPackage = getattr(Multiview, multiviewAlgoName)
mutliviewModule = getattr(multiviewPackage, multiviewAlgoName)
- multiviewArguments += mutliviewModule.getArgs(args, benchmark, views, viewsIndices, randomState, directory, resultsMonoview)
+ multiviewArguments += mutliviewModule.getArgs(args, benchmark, views, viewsIndices, randomState, directory, resultsMonoview, classificationIndices)
argumentDictionaries["Multiview"] = multiviewArguments
return argumentDictionaries
@@ -288,7 +288,7 @@ def classifyOneIter_multicore(LABELS_DICTIONARY, argumentDictionaries, nbCores,
for arguments in argumentDictionaries["Monoview"]]
monoviewTime = time.time() - dataBaseTime - start
- argumentDictionaries = initMultiviewArguments(args, benchmark, views, viewsIndices, argumentDictionaries, randomState, directory, resultsMonoview)
+ argumentDictionaries = initMultiviewArguments(args, benchmark, views, viewsIndices, argumentDictionaries, randomState, directory, resultsMonoview, classificationIndices)
resultsMultiview = []
resultsMultiview += [
@@ -343,10 +343,10 @@ def classifyOneIter(LABELS_DICTIONARY, argumentDictionaries, nbCores, directory,
for arguments in argumentDictionaries["Monoview"]])
monoviewTime = time.time() - dataBaseTime - start
- argumentDictionaries = initMultiviewArguments(args, benchmark, views, viewsIndices, argumentDictionaries, randomState, directory, resultsMonoview)
+ argumentDictionaries = initMultiviewArguments(args, benchmark, views, viewsIndices, argumentDictionaries, randomState, directory, resultsMonoview, classificationIndices)
+ resultsMultiview = []
if nbCores > 1:
- resultsMultiview = []
nbExperiments = len(argumentDictionaries["Multiview"])
for stepIndex in range(int(math.ceil(float(nbExperiments) / nbCores))):
resultsMultiview += Parallel(n_jobs=nbCores)(
@@ -384,6 +384,7 @@ def classifyOneIter(LABELS_DICTIONARY, argumentDictionaries, nbCores, directory,
"s, Multiview Time : "+str(multiviewTime)+
"s, Global Analysis Time : "+str(globalAnalysisTime)+
"s, Total Duration : "+str(totalTime)+"s")
+ return results
def initRandomState(randomStateArg, directory):
@@ -595,6 +596,7 @@ DATASET, LABELS_DICTIONARY = getDatabase(args.views, args.pathF, args.name, args
datasetLength = DATASET.get("Metadata").attrs["datasetLength"]
indices = np.arange(datasetLength)
classificationIndices = genSplits(statsIter, indices, DATASET, args.CL_split, statsIterRandomStates)
+
kFolds = genKFolds(statsIter, args.CL_nbFolds, statsIterRandomStates)
datasetFiles = initMultipleDatasets(args, nbCores)
@@ -631,22 +633,38 @@ argumentDictionaries = initMonoviewArguments(benchmark, argumentDictionaries, vi
directories = genDirecortiesNames(directory, statsIter)
if statsIter>1:
- iterResults = []
- nbExperiments = statsIter
- for stepIndex in range(int(math.ceil(float(nbExperiments) / nbCores))):
- iterResults += (Parallel(n_jobs=nbCores)(
- delayed(classifyOneIter_multicore)(LABELS_DICTIONARY, argumentDictionaries, 1, directories[coreIndex + stepIndex * nbCores], args, classificationIndices[coreIndex + stepIndex * nbCores], kFolds[coreIndex + stepIndex * nbCores],
- statsIterRandomStates[coreIndex + stepIndex * nbCores], hyperParamSearch, metrics, coreIndex, viewsIndices, dataBaseTime, start, benchmark,
- views)
- for coreIndex in range(min(nbCores, nbExperiments - stepIndex * nbCores))))
+ for statIterIndex in range(statsIter):
+ if not os.path.exists(os.path.dirname(directories[statIterIndex]+"train_labels.csv")):
+ try:
+ os.makedirs(os.path.dirname(directories[statIterIndex]+"train_labels.csv"))
+ except OSError as exc:
+ if exc.errno != errno.EEXIST:
+ raise
+ trainIndices, testIndices = classificationIndices[statIterIndex]
+ trainLabels = DATASET.get("Labels").value[trainIndices]
+ np.savetxt(directories[statIterIndex]+"train_labels.csv", trainLabels, delimiter=",")
if nbCores > 1:
+ iterResults = []
+ nbExperiments = statsIter
+ for stepIndex in range(int(math.ceil(float(nbExperiments) / nbCores))):
+ iterResults += (Parallel(n_jobs=nbCores)(
+ delayed(classifyOneIter_multicore)(LABELS_DICTIONARY, argumentDictionaries, 1, directories[coreIndex + stepIndex * nbCores], args, classificationIndices[coreIndex + stepIndex * nbCores], kFolds[coreIndex + stepIndex * nbCores],
+ statsIterRandomStates[coreIndex + stepIndex * nbCores], hyperParamSearch, metrics, coreIndex, viewsIndices, dataBaseTime, start, benchmark,
+ views)
+ for coreIndex in range(min(nbCores, nbExperiments - stepIndex * nbCores))))
logging.debug("Start:\t Deleting " + str(nbCores) + " temporary datasets for multiprocessing")
datasetFiles = DB.deleteHDF5(args.pathF, args.name, nbCores)
logging.debug("Start:\t Deleting datasets for multiprocessing")
+ else:
+ iterResults = []
+ for iterIndex in range(statsIter):
+ iterResults.append(classifyOneIter(LABELS_DICTIONARY, argumentDictionaries, nbCores, directories[iterIndex], args,
+ classificationIndices[iterIndex], kFolds[iterIndex], statsIterRandomStates[iterIndex],
+ hyperParamSearch, metrics, DATASET, viewsIndices, dataBaseTime, start, benchmark, views))
analyzeIterResults(iterResults, args.name, metrics, directory)
else:
- classifyOneIter(LABELS_DICTIONARY, argumentDictionaries, nbCores, directories, args, classificationIndices, kFolds,
+ res = classifyOneIter(LABELS_DICTIONARY, argumentDictionaries, nbCores, directories, args, classificationIndices, kFolds,
statsIterRandomStates, hyperParamSearch, metrics, DATASET, viewsIndices, dataBaseTime, start,
benchmark, views)
diff --git a/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py b/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
index ae0cb5d0..42820970 100644
--- a/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
+++ b/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
@@ -54,7 +54,17 @@ def ExecMonoview(directory, X, Y, name, labelsNames, classificationIndices, KFol
CL_type = kwargs["CL_type"]
X = getValue(X)
learningRate = len(classificationIndices[0])/(len(classificationIndices[0])+len(classificationIndices[1]))
-
+ labelsString = "-".join(labelsNames)
+ timestr = time.strftime("%Y%m%d-%H%M%S")
+ CL_type_string = CL_type
+ outputFileName = directory + "/"+CL_type_string+"/"+"/"+feat+"/"+timestr +"Results-" + CL_type_string + "-" + labelsString + \
+ '-learnRate' + str(learningRate) + '-' + name + "-" + feat + "-"
+ if not os.path.exists(os.path.dirname(outputFileName)):
+ try:
+ os.makedirs(os.path.dirname(outputFileName))
+ except OSError as exc:
+ if exc.errno != errno.EEXIST:
+ raise
logging.debug("Done:\t Loading data")
# Determine the Database to extract features
logging.debug("Info:\t Classification - Database:" + str(name) + " Feature:" + str(feat) + " train ratio:"
@@ -78,7 +88,7 @@ def ExecMonoview(directory, X, Y, name, labelsNames, classificationIndices, KFol
if hyperParamSearch != "None":
classifierHPSearch = getattr(classifierModule, hyperParamSearch)
logging.debug("Start:\t RandomSearch best settings with "+str(nIter)+" iterations for "+CL_type)
- cl_desc = classifierHPSearch(X_train, y_train, randomState, KFolds=KFolds, nbCores=nbCores,
+ cl_desc = classifierHPSearch(X_train, y_train, randomState, outputFileName, KFolds=KFolds, nbCores=nbCores,
metric=metrics[0], nIter=nIter)
clKWARGS = dict((str(index), desc) for index, desc in enumerate(cl_desc))
logging.debug("Done:\t RandomSearch best settings")
@@ -106,17 +116,12 @@ def ExecMonoview(directory, X, Y, name, labelsNames, classificationIndices, KFol
cl_desc = [value for key, value in sorted(clKWARGS.iteritems())]
logging.debug("Done:\t Getting Results")
logging.info(stringAnalysis)
- labelsString = "-".join(labelsNames)
- timestr = time.strftime("%Y%m%d-%H%M%S")
- CL_type_string = CL_type
- outputFileName = directory + "/"+CL_type_string+"/"+"/"+feat+"/"+timestr +"Results-" + CL_type_string + "-" + labelsString + \
- '-learnRate' + str(learningRate) + '-' + name + "-" + feat + "-"
- if not os.path.exists(os.path.dirname(outputFileName)):
- try:
- os.makedirs(os.path.dirname(outputFileName))
- except OSError as exc:
- if exc.errno != errno.EEXIST:
- raise
+ # labelsString = "-".join(labelsNames)
+ # timestr = time.strftime("%Y%m%d-%H%M%S")
+ # CL_type_string = CL_type
+ # outputFileName = directory + "/"+CL_type_string+"/"+"/"+feat+"/"+timestr +"Results-" + CL_type_string + "-" + labelsString + \
+ # '-learnRate' + str(learningRate) + '-' + name + "-" + feat + "-"
+
outputTextFile = open(outputFileName + '.txt', 'w')
outputTextFile.write(stringAnalysis)
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
index 67dab09e..876238ca 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
@@ -4,6 +4,9 @@ from sklearn.model_selection import RandomizedSearchCV
from sklearn.tree import DecisionTreeClassifier
import Metrics
from scipy.stats import randint
+import numpy as np
+import matplotlib.pyplot as plt
+from utils.HyperParameterSearch import genHeatMaps
# Author-Info
__author__ = "Baptiste Bauvin"
@@ -39,10 +42,10 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=4, metric=["accuracy_score", None], nIter=30, nbCores=1):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=4, metric=["accuracy_score", None], nIter=30, nbCores=1):
pipeline = Pipeline([('classifier', AdaBoostClassifier())])
- param= {"classifier__n_estimators": randint(1, 15),
+ param= {"classifier__n_estimators": randint(1, 150),
"classifier__base_estimator": [DecisionTreeClassifier()]}
metricModule = getattr(Metrics, metric[0])
if metric[1]!=None:
@@ -55,6 +58,36 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=4, metric=["accuracy_
detector = grid.fit(X_train, y_train)
desc_estimators = [detector.best_params_["classifier__n_estimators"],
detector.best_params_["classifier__base_estimator"]]
+
+ scoresArray = detector.cv_results_['mean_test_score']
+ params = [("baseEstimators", np.array(["DecisionTree" for _ in range(nIter)])),
+ ("nEstimators", np.array(detector.cv_results_['param_classifier__n_estimators']))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
+
+ # baseEstimatorsSet = np.array(set(baseEstimators))
+ # nEstimatorsSet = np.sort(np.array(list(set(nEstimators))))
+ #
+ # scoresArray = detector.cv_results_['mean_test_score']
+ # scoresMatrix = np.zeros((len(nEstimatorsSet), 1))
+ # for baseEstimator, nEstimator, score in zip(baseEstimators, nEstimators, scoresArray):
+ # baseEstimatorIndex = 0
+ # i, = np.where(nEstimatorsSet == nEstimator)
+ # print i
+ # nEstimatorIndex, = np.where(nEstimatorsSet == nEstimator)
+ # scoresMatrix[int(nEstimatorIndex), baseEstimatorIndex] = score
+ #
+ # plt.figure(figsize=(8, 6))
+ # plt.subplots_adjust(left=.2, right=0.95, bottom=0.15, top=0.95)
+ # plt.imshow(scoresMatrix, interpolation='nearest', cmap=plt.cm.hot,
+ # )
+ # plt.xlabel('n_estimators')
+ # plt.ylabel('base_estimator')
+ # plt.colorbar()
+ # plt.xticks(np.arange(1), ["DecisionTree"])
+ # plt.yticks(np.arange(len(nEstimatorsSet)), nEstimatorsSet, rotation=45)
+ # plt.title('Validation accuracy')
+ # plt.savefig(outputFileName+"heat_map.png")
return desc_estimators
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
index 67acba97..168a0864 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
@@ -3,6 +3,8 @@ from sklearn.pipeline import Pipeline # Pipelining in classifi
from sklearn.model_selection import RandomizedSearchCV
import Metrics
from scipy.stats import randint
+import numpy as np
+from utils.HyperParameterSearch import genHeatMaps
# Author-Info
__author__ = "Baptiste Bauvin"
@@ -42,7 +44,7 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_DT = Pipeline([('classifier', DecisionTreeClassifier())])
param_DT = {"classifier__max_depth": randint(1, 300),
"classifier__criterion": ["gini", "entropy"],
@@ -58,6 +60,13 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=
DT_detector = grid_DT.fit(X_train, y_train)
desc_params = [DT_detector.best_params_["classifier__max_depth"], DT_detector.best_params_["classifier__criterion"],
DT_detector.best_params_["classifier__splitter"]]
+
+ scoresArray = DT_detector.cv_results_['mean_test_score']
+ params = [("maxDepth", np.array(DT_detector.cv_results_['param_classifier__max_depth'])),
+ ("criterion", np.array(DT_detector.cv_results_['param_classifier__criterion'])),
+ ("splitter", np.array(DT_detector.cv_results_['param_classifier__splitter']))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
return desc_params
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py
index c8947a0a..c6df7e41 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py
@@ -3,6 +3,8 @@ from sklearn.pipeline import Pipeline # Pipelining in classifi
from sklearn.model_selection import RandomizedSearchCV
import Metrics
from scipy.stats import randint
+import numpy as np
+from utils.HyperParameterSearch import genHeatMaps
# Author-Info
@@ -47,7 +49,7 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_KNN = Pipeline([('classifier', KNeighborsClassifier())])
param_KNN = {"classifier__n_neighbors": randint(1, 50),
"classifier__weights": ["uniform", "distance"],
@@ -68,6 +70,15 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=
KNN_detector.best_params_["classifier__algorithm"],
KNN_detector.best_params_["classifier__p"],
]
+
+ scoresArray = KNN_detector.cv_results_['mean_test_score']
+ params = [("nNeighbors", np.array(KNN_detector.cv_results_['param_classifier__n_neighbors'])),
+ ("weights", np.array(KNN_detector.cv_results_['param_classifier__weights'])),
+ ("algorithm", np.array(KNN_detector.cv_results_['param_classifier__algorithm'])),
+ ("p", np.array(KNN_detector.cv_results_['param_classifier__p']))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
+
return desc_params
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
index 9452465e..4370d04f 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
@@ -3,6 +3,8 @@ from sklearn.pipeline import Pipeline
from sklearn.model_selection import RandomizedSearchCV
import Metrics
from scipy.stats import randint
+import numpy as np
+from utils.HyperParameterSearch import genHeatMaps
# Author-Info
__author__ = "Baptiste Bauvin"
@@ -43,7 +45,7 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_rf = Pipeline([('classifier', RandomForestClassifier())])
param_rf = {"classifier__n_estimators": randint(1, 300),
"classifier__max_depth": randint(1, 300),
@@ -61,6 +63,13 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=
desc_estimators = [rf_detector.best_params_["classifier__n_estimators"],
rf_detector.best_params_["classifier__max_depth"],
rf_detector.best_params_["classifier__criterion"]]
+
+ scoresArray = rf_detector.cv_results_['mean_test_score']
+ params = [("nEstimators", np.array(rf_detector.cv_results_['param_classifier__n_estimators'])),
+ ("maxDepth", np.array(rf_detector.cv_results_['param_classifier__max_depth'])),
+ ("criterion", np.array(rf_detector.cv_results_['param_classifier__criterion']))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
return desc_estimators
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SCM.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SCM.py
index e5871b66..20d49be8 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SCM.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SCM.py
@@ -6,6 +6,7 @@ import h5py
from Multiview import GetMultiviewDb as DB
from pyscm.binary_attributes.base import BaseBinaryAttributeList
import os
+from utils.HyperParameterSearch import genHeatMaps
# Author-Info
@@ -63,7 +64,7 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=None, metric=["accuracy_score", None], nIter=30, nbCores=1):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=None, metric=["accuracy_score", None], nIter=30, nbCores=1):
metricModule = getattr(Metrics, metric[0])
if metric[1]!=None:
@@ -77,10 +78,16 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=None, metric=["accura
baseScore = 1000.0
isBetter = "lower"
config = []
+ maxAttributesArray = []
+ pArray = []
+ modelsArray = []
for iterIndex in range(nIter):
max_attributes = randomState.randint(1, 20)
+ maxAttributesArray.append(max_attributes)
p = randomState.random_sample()
+ pArray.append(p)
model = randomState.choice(["conjunction", "disjunction"])
+ modelsArray.append(model)
classifier = pyscm.scm.SetCoveringMachine(p=p, max_attributes=max_attributes, model_type=model, verbose=False)
scores = []
kFolds = KFolds.split(X_train, y_train)
@@ -110,6 +117,12 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=None, metric=["accura
config = [max_attributes, p, model]
assert config!=[], "No good configuration found for SCM"
+ scoresArray = scores
+ params = [("maxAttributes", np.array(maxAttributesArray)),
+ ("p", np.array(pArray)),
+ ("model", np.array(modelsArray))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
return config
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
index d531e8e7..4ce68949 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
@@ -3,6 +3,8 @@ from sklearn.pipeline import Pipeline # Pipelining in classifi
from sklearn.model_selection import RandomizedSearchCV
import Metrics
from scipy.stats import uniform
+import numpy as np
+from utils.HyperParameterSearch import genHeatMaps
@@ -47,7 +49,7 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_SGD = Pipeline([('classifier', SGDClassifier())])
losses = ['log', 'modified_huber']
penalties = ["l1", "l2", "elasticnet"]
@@ -65,6 +67,14 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=
SGD_detector = grid_SGD.fit(X_train, y_train)
desc_params = [SGD_detector.best_params_["classifier__loss"], SGD_detector.best_params_["classifier__penalty"],
SGD_detector.best_params_["classifier__alpha"]]
+
+ scoresArray = SGD_detector.cv_results_['mean_test_score']
+ params = [("loss", np.array(SGD_detector.cv_results_['param_classifier__loss'])),
+ ("penalty", np.array(SGD_detector.cv_results_['param_classifier__penalty'])),
+ ("aplha", np.array(SGD_detector.cv_results_['param_classifier__alpha']))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
+
return desc_params
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py
index f9dfb331..0f48fce5 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py
@@ -3,6 +3,8 @@ from sklearn.pipeline import Pipeline # Pipelining in classifi
from sklearn.model_selection import RandomizedSearchCV
import Metrics
from scipy.stats import randint
+import numpy as np
+from utils.HyperParameterSearch import genHeatMaps
# Author-Info
@@ -36,7 +38,7 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_SVMLinear = Pipeline([('classifier', SVC(kernel="linear", max_iter=1000))])
param_SVMLinear = {"classifier__C": randint(1, 10000)}
metricModule = getattr(Metrics, metric[0])
@@ -51,6 +53,13 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=
SVMLinear_detector = grid_SVMLinear.fit(X_train, y_train)
desc_params = [SVMLinear_detector.best_params_["classifier__C"]]
+
+ scoresArray = SVMLinear_detector.cv_results_['mean_test_score']
+ params = [("c", np.array(SVMLinear_detector.cv_results_['param_classifier__C'])),
+ ("control", np.array(["control" for _ in range(nIter)]))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
+
return desc_params
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
index 961e5791..316b7af6 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
@@ -3,6 +3,8 @@ from sklearn.pipeline import Pipeline # Pipelining in classifi
from sklearn.model_selection import RandomizedSearchCV
import Metrics
from scipy.stats import randint
+import numpy as np
+from utils.HyperParameterSearch import genHeatMaps
@@ -40,7 +42,7 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_SVMPoly = Pipeline([('classifier', SVC(kernel="poly", max_iter=1000))])
param_SVMPoly = {"classifier__C": randint(1, 10000),
"classifier__degree": randint(1, 30)}
@@ -52,8 +54,15 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=
scorer = metricModule.get_scorer(**metricKWARGS)
grid_SVMPoly = RandomizedSearchCV(pipeline_SVMPoly, n_iter=nIter, param_distributions=param_SVMPoly, refit=True,
n_jobs=nbCores, scoring=scorer, cv=KFolds, random_state=randomState)
- SVMRBF_detector = grid_SVMPoly.fit(X_train, y_train)
- desc_params = [SVMRBF_detector.best_params_["classifier__C"], SVMRBF_detector.best_params_["classifier__degree"]]
+ SVMPoly_detector = grid_SVMPoly.fit(X_train, y_train)
+ desc_params = [SVMPoly_detector.best_params_["classifier__C"], SVMPoly_detector.best_params_["classifier__degree"]]
+
+ scoresArray = SVMPoly_detector.cv_results_['mean_test_score']
+ params = [("c", np.array(SVMPoly_detector.cv_results_['param_classifier__C'])),
+ ("degree", np.array(SVMPoly_detector.cv_results_['param_classifier__degree']))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
+
return desc_params
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py
index 1ce9881c..4b4ec762 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py
@@ -3,6 +3,8 @@ from sklearn.pipeline import Pipeline # Pipelining in classifi
from sklearn.model_selection import RandomizedSearchCV
import Metrics
from scipy.stats import randint
+import numpy as np
+from utils.HyperParameterSearch import genHeatMaps
# Author-Info
@@ -36,7 +38,7 @@ def getKWARGS(kwargsList):
return kwargsDict
-def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
+def randomizedSearch(X_train, y_train, randomState, outputFileName, KFolds=4, nbCores=1, metric=["accuracy_score", None], nIter=30):
pipeline_SVMRBF = Pipeline([('classifier', SVC(kernel="rbf", max_iter=1000))])
param_SVMRBF = {"classifier__C": randint(1, 10000)}
metricModule = getattr(Metrics, metric[0])
@@ -49,6 +51,13 @@ def randomizedSearch(X_train, y_train, randomState, KFolds=4, nbCores=1, metric=
n_jobs=nbCores, scoring=scorer, cv=KFolds, random_state=randomState)
SVMRBF_detector = grid_SVMRBF.fit(X_train, y_train)
desc_params = [SVMRBF_detector.best_params_["classifier__C"]]
+
+ scoresArray = SVMRBF_detector.cv_results_['mean_test_score']
+ params = [("c", np.array(SVMRBF_detector.cv_results_['param_classifier__C'])),
+ ("control", np.array(["control" for _ in range(nIter)]))]
+
+ genHeatMaps(params, scoresArray, outputFileName)
+
return desc_params
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
index 3370bf1e..8e3012a2 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
@@ -67,7 +67,7 @@ def getBenchmark(benchmark, args=None):
return benchmark
-def getArgs(args, benchmark, views, viewsIndices, randomState, directory, resultsMonoview):
+def getArgs(args, benchmark, views, viewsIndices, randomState, directory, resultsMonoview, classificationIndices):
if not "Monoview" in benchmark and not args.FU_L_select_monoview in ["randomClf", "Determined"]:
args.FU_L_select_monoview = "randomClf"
argumentsList = []
@@ -75,7 +75,7 @@ def getArgs(args, benchmark, views, viewsIndices, randomState, directory, result
fusionTypePackage = getattr(Methods, fusionType+"Package")
for fusionMethod in benchmark["Multiview"]["Fusion"]["Methods"][fusionType]:
fusionMethodModule = getattr(fusionTypePackage, fusionMethod)
- arguments = fusionMethodModule.getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview)
+ arguments = fusionMethodModule.getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview, classificationIndices)
argumentsList+= arguments
return argumentsList
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/EarlyFusionPackage/WeightedLinear.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/EarlyFusionPackage/WeightedLinear.py
index 49e60c71..cd9dbbee 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/EarlyFusionPackage/WeightedLinear.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/EarlyFusionPackage/WeightedLinear.py
@@ -18,7 +18,7 @@ def genParamsSets(classificationKWARGS, randomState, nIter=1):
return paramsSets
-def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview):
+def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview, classificationIndices):
argumentsList = []
if args.FU_E_cl_names != ['']:
pass
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusion.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusion.py
index 1e78855a..09b549cb 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusion.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusion.py
@@ -41,78 +41,103 @@ def getScores(LateFusionClassifiers):
return ""
-def intersect(allClassifersNames, directory, viewsIndices):
- wrongSets = [0 for _ in allClassifersNames]
+def intersect(allClassifersNames, directory, viewsIndices, resultsMonoview, classificationIndices):
+ wrongSets = [[] for _ in viewsIndices]
+ # wrongSets = [0 for _ in allClassifersNames]
+ classifiersNames = [[] for _ in viewsIndices]
nbViews = len(viewsIndices)
- for classifierIndex, classifierName in enumerate(allClassifersNames):
- try :
- classifierDirectory = directory+classifierName+"/"
- viewDirectoryNames = os.listdir(classifierDirectory)
- wrongSets[classifierIndex]=[0 for _ in viewDirectoryNames]
- for viewIndex, viewDirectoryName in enumerate(viewDirectoryNames):
- for resultFileName in os.listdir(classifierDirectory+"/"+viewDirectoryName+"/"):
- if resultFileName.endswith("train_labels.csv"):
- yTrainFileName = classifierDirectory+"/"+viewDirectoryName+"/"+resultFileName
- elif resultFileName.endswith("train_pred.csv"):
- yTrainPredFileName = classifierDirectory+"/"+viewDirectoryName+"/"+resultFileName
- train = np.genfromtxt(yTrainFileName, delimiter=",").astype(np.int16)
- pred = np.genfromtxt(yTrainPredFileName, delimiter=",").astype(np.int16)
- length = len(train)
- wrongLabelsIndices = np.where(train+pred == 1)
- wrongSets[classifierIndex][viewIndex]=wrongLabelsIndices
- except OSError:
- for viewIndex in range(nbViews):
- wrongSets[classifierIndex][viewIndex]= np.arange(length)
- combinations = itertools.combinations_with_replacement(range(len(allClassifersNames)), nbViews)
+ trainLabels = np.genfromtxt(directory+"train_labels.csv", delimiter=",").astype(np.int16)
+ length = len(trainLabels)
+ for resultMonoview in resultsMonoview:
+ if resultMonoview[1][0] in classifiersNames[resultMonoview[0]]:
+ classifierIndex = classifiersNames.index(resultMonoview[1][0])
+ wrongSets[resultMonoview[0]][classifierIndex] = np.where(trainLabels+resultMonoview[1][3][classificationIndices[0]] == 1)
+ else:
+ classifiersNames[resultMonoview[0]].append(resultMonoview[1][0])
+ wrongSets[resultMonoview[0]].append(np.where(trainLabels+resultMonoview[1][3][classificationIndices[0]] == 1))
+ # for classifierIndex, classifierName in enumerate(allClassifersNames):
+ # try:
+ # classifierDirectory = directory+classifierName+"/"
+ # viewDirectoryNames = os.listdir(classifierDirectory)
+ # wrongSets[classifierIndex]=[0 for _ in viewDirectoryNames]
+ # for viewIndex, viewDirectoryName in enumerate(viewDirectoryNames):
+ # for resultFileName in os.listdir(classifierDirectory+"/"+viewDirectoryName+"/"):
+ # if resultFileName.endswith("train_labels.csv"):
+ # yTrainFileName = classifierDirectory+"/"+viewDirectoryName+"/"+resultFileName
+ # elif resultFileName.endswith("train_pred.csv"):
+ # yTrainPredFileName = classifierDirectory+"/"+viewDirectoryName+"/"+resultFileName
+ # train = np.genfromtxt(yTrainFileName, delimiter=",").astype(np.int16)
+ # pred = np.genfromtxt(yTrainPredFileName, delimiter=",").astype(np.int16)
+ # length = len(train)
+ # wrongLabelsIndices = np.where(train+pred == 1)
+ # wrongSets[classifierIndex][viewIndex]=wrongLabelsIndices
+ # except OSError:
+ # for viewIndex in range(nbViews):
+ # wrongSets[classifierIndex][viewIndex]= np.arange(length)
+
+ combinations = itertools.combinations_with_replacement(range(len(classifiersNames[0])), nbViews)
bestLen = length
bestCombination = None
for combination in combinations:
intersect = np.arange(length, dtype=np.int16)
for viewIndex, classifierIndex in enumerate(combination):
- intersect = np.intersect1d(intersect, wrongSets[classifierIndex][viewIndex])
+ intersect = np.intersect1d(intersect, wrongSets[viewIndex][classifierIndex])
if len(intersect) < bestLen:
bestLen = len(intersect)
bestCombination = combination
- return [allClassifersNames[index] for index in bestCombination]
-
-
-def getFormFile(directory, viewDirectory, resultFileName):
- file = open(directory+"/"+viewDirectory+"/"+resultFileName)
- for line in file:
- if "Score on train" in line:
- score = float(line.strip().split(":")[1])
- break
- elif "train" in line:
- metricName = line.strip().split(" ")[0]
+ return [classifiersNames[viewIndex][index] for viewIndex, index in enumerate(bestCombination)]
+
+
+# def getFormFile(directory, viewDirectory, resultFileName):
+# file = open(directory+"/"+viewDirectory+"/"+resultFileName)
+# for line in file:
+# if "Score on train" in line:
+# score = float(line.strip().split(":")[1])
+# break
+# elif "train" in line:
+# metricName = line.strip().split(" ")[0]
+# metricModule = getattr(Metrics, metricName)
+# if metricModule.getConfig()[-14]=="h":
+# betterHigh = True
+# else:
+# betterHigh = False
+# return score, betterHigh
+
+
+def bestScore(allClassifersNames, directory, viewsIndices, resultsMonoview, classificationIndices):
+ nbViews = len(viewsIndices)
+ nbClassifiers = len(allClassifersNames)
+ scores = np.zeros((nbViews, nbClassifiers))
+ classifiersNames = [[] for _ in viewsIndices]
+ metricName = resultsMonoview[0][1][2].keys()[0]
metricModule = getattr(Metrics, metricName)
if metricModule.getConfig()[-14]=="h":
betterHigh = True
else:
betterHigh = False
- return score, betterHigh
-
-
-def bestScore(allClassifersNames, directory, viewsIndices):
- nbViews = len(viewsIndices)
- nbClassifiers = len(allClassifersNames)
- scores = np.zeros((nbViews, nbClassifiers))
- for classifierIndex, classifierName in enumerate(allClassifersNames):
- classifierDirectory = directory+"/"+classifierName+"/"
- for viewIndex, viewDirectory in enumerate(os.listdir(classifierDirectory)):
- for resultFileName in os.listdir(classifierDirectory+"/"+viewDirectory+"/"):
- if resultFileName.endswith(".txt"):
- scores[viewIndex, classifierIndex], betterHigh = getFormFile(directory, viewDirectory, resultFileName)
+ for resultMonoview in resultsMonoview:
+ if resultMonoview[1][0] not in classifiersNames[resultMonoview[0]]:
+ classifiersNames[resultMonoview[0]].append(resultMonoview[1][0])
+ classifierIndex = classifiersNames[resultMonoview[0]].index(resultMonoview[1][0])
+ scores[resultMonoview[0],classifierIndex] = resultMonoview[1][2].values()[0][0]
+ #
+ # for classifierIndex, classifierName in enumerate(allClassifersNames):
+ # classifierDirectory = directory+"/"+classifierName+"/"
+ # for viewIndex, viewDirectory in enumerate(os.listdir(classifierDirectory)):
+ # for resultFileName in os.listdir(classifierDirectory+"/"+viewDirectory+"/"):
+ # if resultFileName.endswith(".txt"):
+ # scores[viewIndex, classifierIndex], betterHigh = getFormFile(directory, viewDirectory, resultFileName)
if betterHigh:
classifierIndices = np.argmax(scores, axis=1)
else:
classifierIndices = np.argmin(scores, axis=1)
- return [allClassifersNames[index] for index in classifierIndices]
+ return [classifiersNames[viewIndex][index] for viewIndex, index in enumerate(classifierIndices)]
-def getClassifiers(selectionMethodName, allClassifiersNames, directory, viewsIndices):
+def getClassifiers(selectionMethodName, allClassifiersNames, directory, viewsIndices, resultsMonoview, classificationIndices):
thismodule = sys.modules[__name__]
selectionMethod = getattr(thismodule, selectionMethodName)
- classifiersNames = selectionMethod(allClassifiersNames, directory, viewsIndices)
+ classifiersNames = selectionMethod(allClassifiersNames, directory, viewsIndices, resultsMonoview, classificationIndices)
return classifiersNames
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/BayesianInference.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/BayesianInference.py
index fd1bc24f..32d623ed 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/BayesianInference.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/BayesianInference.py
@@ -17,12 +17,12 @@ def genParamsSets(classificationKWARGS, randomState, nIter=1):
return paramsSets
-def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview):
+def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview, classificationIndices):
if args.FU_L_cl_names!=['']:
args.FU_L_select_monoview = "user_defined"
else:
monoviewClassifierModulesNames = benchmark["Monoview"]
- args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices)
+ args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices, resultsMonoview, classificationIndices)
monoviewClassifierModules = [getattr(MonoviewClassifiers, classifierName)
for classifierName in args.FU_L_cl_names]
if args.FU_L_cl_names==[""] and args.CL_type == ["Multiview"]:
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/MajorityVoting.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/MajorityVoting.py
index 19e2f400..31d44c79 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/MajorityVoting.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/MajorityVoting.py
@@ -16,12 +16,12 @@ def genParamsSets(classificationKWARGS, randomState, nIter=1):
return paramsSets
-def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview):
+def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview, classificationIndices):
if args.FU_L_cl_names!=['']:
pass
else:
monoviewClassifierModulesNames = benchmark["Monoview"]
- args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices)
+ args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices, resultsMonoview, classificationIndices)
monoviewClassifierModules = [getattr(MonoviewClassifiers, classifierName)
for classifierName in args.FU_L_cl_names]
if args.FU_L_cl_config != ['']:
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/SCMForLinear.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/SCMForLinear.py
index e675b502..5ba55d97 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/SCMForLinear.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/SCMForLinear.py
@@ -27,12 +27,12 @@ def genParamsSets(classificationKWARGS, randomState, nIter=1):
return paramsSets
-def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview):
+def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview, classificationIndices):
if args.FU_L_cl_names!=['']:
pass
else:
monoviewClassifierModulesNames =benchmark["Monoview"]
- args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices)
+ args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices, resultsMonoview, classificationIndices)
monoviewClassifierModules = [getattr(MonoviewClassifiers, classifierName)
for classifierName in args.FU_L_cl_names]
if args.FU_L_cl_config != ['']:
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/SVMForLinear.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/SVMForLinear.py
index ebad066f..d488d7c1 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/SVMForLinear.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/SVMForLinear.py
@@ -14,12 +14,12 @@ def genParamsSets(classificationKWARGS, randomState, nIter=1):
return paramsSets
-def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview):
+def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview, classificationIndices):
if args.FU_L_cl_names!=['']:
pass
else:
monoviewClassifierModulesNames = benchmark["Monoview"]
- args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices)
+ args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices, resultsMonoview, classificationIndices)
monoviewClassifierModules = [getattr(MonoviewClassifiers, classifierName)
for classifierName in args.FU_L_cl_names]
if args.FU_L_cl_config != ['']:
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/WeightedLinear.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/WeightedLinear.py
index eb9678e7..20c16340 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/WeightedLinear.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/LateFusionPackage/WeightedLinear.py
@@ -16,12 +16,12 @@ def genParamsSets(classificationKWARGS, randomState, nIter=1):
return paramsSets
-def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview):
+def getArgs(benchmark, args, views, viewsIndices, directory, resultsMonoview, classificationIndices):
if args.FU_L_cl_names!=['']:
pass
else:
monoviewClassifierModulesNames = benchmark["Monoview"]
- args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices)
+ args.FU_L_cl_names = getClassifiers(args.FU_L_select_monoview, monoviewClassifierModulesNames, directory, viewsIndices, resultsMonoview, classificationIndices)
monoviewClassifierModules = [getattr(MonoviewClassifiers, classifierName)
for classifierName in args.FU_L_cl_names]
if args.FU_L_cl_config != ['']:
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
index 3b63701d..e5ae0318 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
@@ -28,7 +28,7 @@ def getBenchmark(benchmark, args=None):
return benchmark
-def getArgs(args, benchmark, views, viewsIndices, randomState, directory, resultsMonoview):
+def getArgs(args, benchmark, views, viewsIndices, randomState, directory, resultsMonoview, classificationIndices):
argumentsList = []
arguments = {"CL_type": "Mumbo",
diff --git a/Code/MonoMutliViewClassifiers/ResultAnalysis.py b/Code/MonoMutliViewClassifiers/ResultAnalysis.py
index 0cce2e35..80a577b6 100644
--- a/Code/MonoMutliViewClassifiers/ResultAnalysis.py
+++ b/Code/MonoMutliViewClassifiers/ResultAnalysis.py
@@ -78,6 +78,7 @@ def resultAnalysis(benchmark, results, name, times, metrics, directory):
ax.set_xticklabels(names, rotation="vertical")
f.savefig(directory+time.strftime("%Y%m%d-%H%M%S")+"-"+name+"-"+metric[0]+".png")
+ plt.close()
def analyzeLabels(labelsArrays, realLabels, results, directory):
@@ -104,6 +105,7 @@ def analyzeLabels(labelsArrays, realLabels, results, directory):
cbar = fig.colorbar(cax, ticks=[0, 1])
cbar.ax.set_yticklabels(['Wrong', ' Right'])
fig.savefig(directory+time.strftime("%Y%m%d-%H%M%S")+"-error_analysis.png")
+ plt.close()
def genScoresNames(iterResults, metric, nbResults, names, nbMono):
@@ -154,7 +156,6 @@ def genScoresNames(iterResults, metric, nbResults, names, nbMono):
def analyzeIterResults(iterResults, name, metrics, directory):
nbResults = len(iterResults[0][0])+len(iterResults[0][1])
nbMono = len(iterResults[0][0])
- nbMulti = len(iterResults[0][1])
nbIter = len(iterResults)
names = genNamesFromRes(iterResults[0][0], iterResults[0][1])
for metric in metrics:
diff --git a/Code/MonoMutliViewClassifiers/utils/HyperParameterSearch.py b/Code/MonoMutliViewClassifiers/utils/HyperParameterSearch.py
index 573c7388..0f98ef52 100644
--- a/Code/MonoMutliViewClassifiers/utils/HyperParameterSearch.py
+++ b/Code/MonoMutliViewClassifiers/utils/HyperParameterSearch.py
@@ -3,6 +3,8 @@ import numpy as np
import sys
import Multiview
import Metrics
+import matplotlib.pyplot as plt
+import itertools
def searchBestSettings(dataset, classifierName, metrics, iLearningIndices, iKFolds, randomState, viewsIndices=None, searchingTool="hyperParamSearch", nIter=1, **kwargs):
if viewsIndices is None:
@@ -71,6 +73,39 @@ def randomizedSearch(dataset, classifierName, metrics, learningIndices, KFolds,
def spearMint(dataset, classifierName, viewsIndices=None, kFolds=None, nIter=1, **kwargs):
pass
+
+def genHeatMaps(params, scoresArray, outputFileName):
+ nbParams = len(params)
+ if nbParams > 2:
+ combinations = itertools.combinations(range(nbParams), 2)
+ else:
+ combinations = [(0,1)]
+ for combination in combinations:
+ paramName1, paramArray1 = params[combination[0]]
+ paramName2, paramArray2 = params[combination[1]]
+
+ paramArray1Set = np.sort(np.array(list(set(paramArray1))))
+ paramArray2Set = np.sort(np.array(list(set(paramArray2))))
+
+ scoresMatrix = np.zeros((len(paramArray2Set), len(paramArray1Set)))-0.1
+ for param1, param2, score in zip(paramArray1, paramArray2, scoresArray):
+ param1Index, = np.where(paramArray1Set == param1)
+ param2Index, = np.where(paramArray2Set == param2)
+ scoresMatrix[int(param2Index), int(param1Index)] = score
+
+ plt.figure(figsize=(8, 6))
+ plt.subplots_adjust(left=.2, right=0.95, bottom=0.15, top=0.95)
+ plt.imshow(scoresMatrix, interpolation='nearest', cmap=plt.cm.hot,
+ )
+ plt.xlabel(paramName1)
+ plt.ylabel(paramName2)
+ plt.colorbar()
+ plt.xticks(np.arange(len(paramArray1Set)), paramArray1Set)
+ plt.yticks(np.arange(len(paramArray2Set)), paramArray2Set, rotation=45)
+ plt.title('Validation metric')
+ plt.savefig(outputFileName+"heat_map-"+paramName1+"-"+paramName2+".png")
+ plt.close()
+
# nohup python ~/dev/git/spearmint/spearmint/main.py . &
# import json
--
GitLab