From 6d384da476d8de9d570d71fc5ecaf19007dca71b Mon Sep 17 00:00:00 2001
From: bbauvin <baptiste.bauvin@centrale-marseille.fr>
Date: Mon, 29 Aug 2016 10:08:30 -0400
Subject: [PATCH] Merged gridsearches and added custom metrics for gs need to
add it to result analysis
---
Code/MonoMutliViewClassifiers/ExecClassif.py | 8 +-
.../Metrics/accuracy_score.py | 9 ++
.../Metrics/f1_score.py | 22 +++++
.../Metrics/fbeta_score.py | 26 ++++++
.../Metrics/hamming_loss.py | 9 ++
.../Metrics/jaccard_similarity_score.py | 9 ++
.../Metrics/log_loss.py | 13 +++
.../Metrics/matthews_corrcoef.py | 7 +-
.../Metrics/precision_score.py | 22 +++++
.../Metrics/recall_score.py | 22 +++++
.../Metrics/roc_auc_score.py | 13 +++
.../Metrics/zero_one_loss.py | 9 ++
.../Monoview/ExecClassifMonoView.py | 15 ++-
.../MonoviewClassifiers/Adaboost.py | 31 ++++---
.../MonoviewClassifiers/DecisionTree.py | 27 +++---
.../MonoviewClassifiers/KNN.py | 25 +++--
.../MonoviewClassifiers/RandomForest.py | 68 +++++++-------
.../MonoviewClassifiers/SGD.py | 29 +++---
.../MonoviewClassifiers/SVMLinear.py | 26 ++++--
.../MonoviewClassifiers/SVMPoly.py | 31 +++++--
.../MonoviewClassifiers/SVMRBF.py | 25 +++--
.../Multiview/ExecMultiview.py | 8 +-
.../Multiview/Fusion/Fusion.py | 26 +++++-
.../Multiview/Fusion/Methods/EarlyFusion.py | 92 +++++++++----------
.../Mumbo/Classifiers/DecisionTree.py | 2 +-
.../Multiview/Mumbo/Classifiers/Kover.py | 7 +-
.../Multiview/Mumbo/Mumbo.py | 12 +--
27 files changed, 411 insertions(+), 182 deletions(-)
diff --git a/Code/MonoMutliViewClassifiers/ExecClassif.py b/Code/MonoMutliViewClassifiers/ExecClassif.py
index 4f52c31c..54b27ba1 100644
--- a/Code/MonoMutliViewClassifiers/ExecClassif.py
+++ b/Code/MonoMutliViewClassifiers/ExecClassif.py
@@ -58,6 +58,8 @@ groupClass.add_argument('--CL_algos_multiview', metavar='STRING', action='store'
help='Determine which multiview classifier to use, separate with : if multiple, if empty, considering all', default='')
groupClass.add_argument('--CL_cores', metavar='INT', action='store', help='Number of cores, -1 for all', type=int,
default=1)
+groupClass.add_argument('--CL_metrics', metavar='STRING', action='store',
+ help='Determine which metric to use, separate with ":" if multiple, if empty, considering all', default='')
groupRF = parser.add_argument_group('Random Forest arguments')
groupRF.add_argument('--CL_RF_trees', metavar='STRING', action='store', help='GridSearch: Determine the trees',
@@ -198,9 +200,9 @@ if "Monoview" in args.CL_type.strip(":"):
fusionClassifierConfig = "a"
fusionMethodConfig = "a"
-mumboNB_ITER = 2
mumboClassifierConfig = "a"
mumboclassifierNames = "a"
+metrics = args.CL_metrics.split(":")
RandomForestKWARGS = {"0":map(int, args.CL_RF_trees.split())}
SVMLinearKWARGS = {"0":map(int, args.CL_SVML_C.split(":"))}
@@ -230,7 +232,7 @@ for viewIndex, viewArguments in enumerate(argumentDictionaries["Monoview"].value
resultsMonoview += (Parallel(n_jobs=nbCores)(
delayed(ExecMonoview)(DATASET.get("View"+str(viewIndex)), DATASET.get("labels").value, args.name,
args.CL_split, args.CL_nbFolds, 1, args.type, args.pathF, gridSearch=True,
- **arguments)
+ metrics=metrics[viewIndex], **arguments)
for arguments in viewArguments))
accuracies = [result[1] for result in resultsMonoview[viewIndex]]
@@ -283,7 +285,7 @@ for viewIndex, viewArguments in enumerate(argumentDictionaries["Monoview"].value
# resultsMultiview = Parallel(n_jobs=nbCores)(
# delayed(ExecMultiview)(DATASET, args.name, args.CL_split, args.CL_nbFolds, 1, args.type, args.pathF,
-# LABELS_DICTIONARY, gridSearch=True, **arguments)
+# LABELS_DICTIONARY, gridSearch=True, metrics=metrics, **arguments)
# for arguments in argumentDictionaries["Multiview"])
resultsMultiview = []
results = (resultsMonoview, resultsMultiview)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/accuracy_score.py b/Code/MonoMutliViewClassifiers/Metrics/accuracy_score.py
index 08d5c30d..226326f9 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/accuracy_score.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/accuracy_score.py
@@ -1,4 +1,5 @@
from sklearn.metrics import accuracy_score as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -8,3 +9,11 @@ def score(y_true, y_pred, **kwargs):
sample_weight=None
score = metric(y_true, y_pred, sample_weight=sample_weight)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight=None
+ return make_scorer(metric, greater_is_better=True, sample_weight=sample_weight)
\ No newline at end of file
diff --git a/Code/MonoMutliViewClassifiers/Metrics/f1_score.py b/Code/MonoMutliViewClassifiers/Metrics/f1_score.py
index 3dc8c3ae..d563d7dc 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/f1_score.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/f1_score.py
@@ -1,4 +1,5 @@
from sklearn.metrics import f1_score as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -20,3 +21,24 @@ def score(y_true, y_pred, **kwargs):
average = "binary"
score = metric(y_true, y_pred, sample_weight=sample_weight, labels=labels, pos_label=pos_label, average=average)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight=None
+ try:
+ labels = kwargs["1"]
+ except:
+ labels=None
+ try:
+ pos_label = kwargs["2"]
+ except:
+ pos_label = 1
+ try:
+ average = kwargs["3"]
+ except:
+ average = "binary"
+ return make_scorer(metric, greater_is_better=True, sample_weight=sample_weight, labels=labels,
+ pos_label=pos_label, average=average)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/fbeta_score.py b/Code/MonoMutliViewClassifiers/Metrics/fbeta_score.py
index 74f1e241..73d834e2 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/fbeta_score.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/fbeta_score.py
@@ -1,4 +1,5 @@
from sklearn.metrics import fbeta_score as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -24,3 +25,28 @@ def score(y_true, y_pred, **kwargs):
average = "binary"
score = metric(y_true, y_pred, beta, sample_weight=sample_weight, labels=labels, pos_label=pos_label, average=average)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight=None
+ try:
+ beta = kwargs["1"]
+ except:
+ beta=1.0
+ try:
+ labels = kwargs["2"]
+ except:
+ labels=None
+ try:
+ pos_label = kwargs["3"]
+ except:
+ pos_label = 1
+ try:
+ average = kwargs["4"]
+ except:
+ average = "binary"
+ return make_scorer(metric, greater_is_better=True, beta=beta, sample_weight=sample_weight, labels=labels,
+ pos_label=pos_label, average=average)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/hamming_loss.py b/Code/MonoMutliViewClassifiers/Metrics/hamming_loss.py
index fe35066d..8bc33bc2 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/hamming_loss.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/hamming_loss.py
@@ -1,4 +1,5 @@
from sklearn.metrics import hamming_loss as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -8,3 +9,11 @@ def score(y_true, y_pred, **kwargs):
classes=None
score = metric(y_true, y_pred, classes=classes)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ classes = kwargs["0"]
+ except:
+ classes=None
+ return make_scorer(metric, greater_is_better=False, classes=classes)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/jaccard_similarity_score.py b/Code/MonoMutliViewClassifiers/Metrics/jaccard_similarity_score.py
index c6d134ad..18e0b484 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/jaccard_similarity_score.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/jaccard_similarity_score.py
@@ -1,4 +1,5 @@
from sklearn.metrics import jaccard_similarity_score as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -8,3 +9,11 @@ def score(y_true, y_pred, **kwargs):
sample_weight = None
score = metric(y_true, y_pred, sample_weight=sample_weight)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight = None
+ return make_scorer(metric, greater_is_better=True, sample_weight=sample_weight)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/log_loss.py b/Code/MonoMutliViewClassifiers/Metrics/log_loss.py
index a6dbf8a3..cb14bee9 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/log_loss.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/log_loss.py
@@ -1,4 +1,5 @@
from sklearn.metrics import log_loss as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -12,3 +13,15 @@ def score(y_true, y_pred, **kwargs):
eps = 1e-15
score = metric(y_true, y_pred, sample_weight=sample_weight, eps=eps)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight = None
+ try:
+ eps = kwargs["1"]
+ except:
+ eps = 1e-15
+ return make_scorer(metric, greater_is_better=False, sample_weight=sample_weight, eps=eps)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/matthews_corrcoef.py b/Code/MonoMutliViewClassifiers/Metrics/matthews_corrcoef.py
index b1ec857a..2a3c4447 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/matthews_corrcoef.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/matthews_corrcoef.py
@@ -1,6 +1,11 @@
from sklearn.metrics import matthews_corrcoef as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
score = metric(y_true, y_pred)
- return score
\ No newline at end of file
+ return score
+
+
+def get_scorer(**kwargs):
+ return make_scorer(metric, greater_is_better=True)
\ No newline at end of file
diff --git a/Code/MonoMutliViewClassifiers/Metrics/precision_score.py b/Code/MonoMutliViewClassifiers/Metrics/precision_score.py
index 2a6cccad..9839c817 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/precision_score.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/precision_score.py
@@ -1,4 +1,5 @@
from sklearn.metrics import precision_score as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -20,3 +21,24 @@ def score(y_true, y_pred, **kwargs):
average = "binary"
score = metric(y_true, y_pred, sample_weight=sample_weight, labels=labels, pos_label=pos_label, average=average)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight=None
+ try:
+ labels = kwargs["1"]
+ except:
+ labels=None
+ try:
+ pos_label = kwargs["2"]
+ except:
+ pos_label = 1
+ try:
+ average = kwargs["3"]
+ except:
+ average = "binary"
+ return make_scorer(metric, greater_is_better=True, sample_weight=sample_weight, labels=labels, pos_label=pos_label,
+ average=average)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/recall_score.py b/Code/MonoMutliViewClassifiers/Metrics/recall_score.py
index a4d10284..f4d38465 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/recall_score.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/recall_score.py
@@ -1,4 +1,5 @@
from sklearn.metrics import recall_score as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -20,3 +21,24 @@ def score(y_true, y_pred, **kwargs):
average = "binary"
score = metric(y_true, y_pred, sample_weight=sample_weight, labels=labels, pos_label=pos_label, average=average)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight=None
+ try:
+ labels = kwargs["1"]
+ except:
+ labels=None
+ try:
+ pos_label = kwargs["2"]
+ except:
+ pos_label = 1
+ try:
+ average = kwargs["3"]
+ except:
+ average = "binary"
+ return make_scorer(metric, greater_is_better=True, sample_weight=sample_weight, labels=labels, pos_label=pos_label,
+ average=average)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/roc_auc_score.py b/Code/MonoMutliViewClassifiers/Metrics/roc_auc_score.py
index 5a9f7cfb..04fb96cb 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/roc_auc_score.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/roc_auc_score.py
@@ -1,4 +1,5 @@
from sklearn.metrics import roc_auc_score as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -12,3 +13,15 @@ def score(y_true, y_pred, **kwargs):
average = "binary"
score = metric(y_true, y_pred, sample_weight=sample_weight, average=average)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight=None
+ try:
+ average = kwargs["1"]
+ except:
+ average = "binary"
+ return make_scorer(metric, greater_is_better=True, sample_weight=sample_weight, average=average)
diff --git a/Code/MonoMutliViewClassifiers/Metrics/zero_one_loss.py b/Code/MonoMutliViewClassifiers/Metrics/zero_one_loss.py
index a741825b..833d2402 100644
--- a/Code/MonoMutliViewClassifiers/Metrics/zero_one_loss.py
+++ b/Code/MonoMutliViewClassifiers/Metrics/zero_one_loss.py
@@ -1,4 +1,5 @@
from sklearn.metrics import zero_one_loss as metric
+from sklearn.metrics import make_scorer
def score(y_true, y_pred, **kwargs):
@@ -8,3 +9,11 @@ def score(y_true, y_pred, **kwargs):
sample_weight=None
score = metric(y_true, y_pred, sample_weight=sample_weight)
return score
+
+
+def get_scorer(**kwargs):
+ try:
+ sample_weight = kwargs["0"]
+ except:
+ sample_weight=None
+ return make_scorer(metric, greater_is_better=False, sample_weight=sample_weight)
diff --git a/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py b/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
index dda9e5d7..9e2a187c 100644
--- a/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
+++ b/Code/MonoMutliViewClassifiers/Monoview/ExecClassifMonoView.py
@@ -20,6 +20,7 @@ import h5py
import ClassifMonoView # Functions for classification
import ExportResults # Functions to render results
import MonoviewClassifiers
+import Metrics
# Author-Info
__author__ = "Nikolas Huelsmann, Baptiste BAUVIN"
@@ -30,7 +31,8 @@ __date__ = 2016-03-25
### Argument Parser
-def ExecMonoview(X, Y, name, learningRate, nbFolds, nbCores, databaseType, path, gridSearch=True, **kwargs):
+def ExecMonoview(X, Y, name, learningRate, nbFolds, nbCores, databaseType, path, gridSearch=True,
+ metrics="accuracy_score", **kwargs):
t_start = time.time()
directory = os.path.dirname(os.path.abspath(__file__)) + "/Results-ClassMonoView/"
@@ -41,6 +43,7 @@ def ExecMonoview(X, Y, name, learningRate, nbFolds, nbCores, databaseType, path,
CL_type = kwargs["CL_type"]
classifierKWARGS = kwargs[CL_type+"KWARGS"]
X = X.value
+ metrics = [getattr(Metrics, metric) for metric in metrics]
# Determine the Database to extract features
logging.debug("### Main Programm for Classification MonoView")
@@ -61,9 +64,10 @@ def ExecMonoview(X, Y, name, learningRate, nbFolds, nbCores, databaseType, path,
classifierModule = getattr(MonoviewClassifiers, CL_type)
- classifierFunction = getattr(classifierModule, "fit_gridsearch")
+ classifierGridSearch = getattr(classifierModule, "gridSearch")
- cl_desc, cl_res = classifierFunction(X_train, y_train, nbFolds=nbFolds, nbCores=nbCores,**classifierKWARGS)
+ cl_desc = classifierGridSearch(X_train, y_train, nbFolds=nbFolds, nbCores=nbCores, metrics=metrics)
+ cl_res = classifierModule.fit(X_train, y_train, NB_CORES=nbCores)
t_end = time.time() - t_start
# Add result to Results DF
@@ -149,6 +153,8 @@ if __name__=='__main__':
groupClass.add_argument('--CL_CV', metavar='INT', action='store', help='Number of k-folds for CV', type=int, default=10)
groupClass.add_argument('--CL_Cores', metavar='INT', action='store', help='Number of cores, -1 for all', type=int, default=1)
groupClass.add_argument('--CL_split', metavar='FLOAT', action='store', help='Split ratio for train and test', type=float, default=0.9)
+ groupClass.add_argument('--CL_metrics', metavar='STRING', action='store',
+ help='Determine which metric to use, separate with ":" if multiple, if empty, considering all', default='')
groupClassifier = parser.add_argument_group('Classifier Config')
@@ -217,4 +223,5 @@ if __name__=='__main__':
arguments = {args.CL_type+"KWARGS": classifierKWARGS, "feat":args.feat,"fileFeat": args.fileFeat,
"fileCL": args.fileCL, "fileCLD": args.fileCLD, "CL_type": args.CL_type}
- ExecMonoview(X, Y, args.name, args.CL_split, args.CL_CV, args.CL_Cores, args.type, args.pathF, **arguments)
+ ExecMonoview(X, Y, args.name, args.CL_split, args.CL_CV, args.CL_Cores, args.type, args.pathF,
+ metrics=args.CL_metrics, **arguments)
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
index 6eba0119..df9269d6 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/Adaboost.py
@@ -5,6 +5,7 @@ from sklearn.tree import DecisionTreeClassifier
from sklearn.utils.testing import all_estimators
import inspect
import numpy as np
+import Metrics
def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
@@ -14,25 +15,29 @@ def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
classifier.fit(DATASET, CLASS_LABELS)
return "No desc", classifier
-
-def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
- pipeline = Pipeline([('classifier', AdaBoostClassifier())])
- param= {"classifier__n_estimators": map(int, kwargs['0']),
- "classifier__base_estimator": [DecisionTreeClassifier() for arg in kwargs["1"]]}
- grid = RandomizedSearchCV(pipeline,param_distributions=param,refit=True,n_jobs=nbCores,scoring='accuracy',cv=nbFolds)
- detector = grid.fit(X_train, y_train)
- desc_estimators = [detector.best_params_["classifier__n_estimators"]]
- description = "Classif_" + "RF" + "-" + "CV_" + str(nbFolds) + "-" + "Trees_" + str(map(str,desc_estimators))
- return description, detector
-
-
-def gridSearch(X_train, y_train, nbFolds=4, nbCores=1):
+#
+# def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+# pipeline = Pipeline([('classifier', AdaBoostClassifier())])
+# metricModule = getattr(Metrics, metric[0])
+# scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
+# param= {"classifier__n_estimators": map(int, kwargs['0']),
+# "classifier__base_estimator": [DecisionTreeClassifier() for arg in kwargs["1"]]}
+# grid = RandomizedSearchCV(pipeline,param_distributions=param,refit=True,n_jobs=nbCores,scoring=scorer,cv=nbFolds)
+# detector = grid.fit(X_train, y_train)
+# desc_estimators = [detector.best_params_["classifier__n_estimators"]]
+# description = "Classif_" + "RF" + "-" + "CV_" + str(nbFolds) + "-" + "Trees_" + str(map(str,desc_estimators))
+# return description, detector
+
+
+def gridSearch(X_train, y_train, nbFolds=4, metric=["accuracy_score", None], nbCores=1):
pipeline = Pipeline([('classifier', AdaBoostClassifier())])
classifiers = [clf for name, clf in all_estimators(type_filter='classifier')
if 'sample_weight' in inspect.getargspec(clf().fit)[0]
and (name != "AdaBoostClassifier" and name !="GradientBoostingClassifier")]
param= {"classifier__n_estimators": np.random.randint(1, 30, 10),
"classifier__base_estimator": classifiers}
+ metricModule = getattr(Metrics, metric[0])
+ scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
grid = RandomizedSearchCV(pipeline,param_distributions=param,refit=True,n_jobs=nbCores,scoring='accuracy',cv=nbFolds)
detector = grid.fit(X_train, y_train)
desc_estimators = [detector.best_params_["classifier__n_estimators"],
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
index c9c86c35..ce7e739b 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/DecisionTree.py
@@ -2,6 +2,7 @@ from sklearn.tree import DecisionTreeClassifier
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.grid_search import GridSearchCV
import numpy as np
+import Metrics
def fit(DATASET, CLASS_LABELS, NB_CORES=1, **kwargs):
maxDepth = int(kwargs['0'])
@@ -10,22 +11,24 @@ def fit(DATASET, CLASS_LABELS, NB_CORES=1, **kwargs):
return "No desc", classifier
-def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
- pipeline_DT = Pipeline([('classifier', DecisionTreeClassifier())])
- param_DT = {"classifier__max_depth":map(int, kwargs['0'])}
-
- grid_DT = GridSearchCV(pipeline_DT, param_grid=param_DT, refit=True, n_jobs=nbCores, scoring='accuracy',
- cv=nbFolds)
- DT_detector = grid_DT.fit(X_train, y_train)
- desc_params = [DT_detector.best_params_["classifier__max_depth"]]
- description = "Classif_" + "DT" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
- return description, DT_detector
+# def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+# pipeline_DT = Pipeline([('classifier', DecisionTreeClassifier())])
+# param_DT = {"classifier__max_depth":map(int, kwargs['0'])}
+# metricModule = getattr(Metrics, metric[0])
+# scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
+# grid_DT = GridSearchCV(pipeline_DT, param_grid=param_DT, refit=True, n_jobs=nbCores, scoring='accuracy',
+# cv=nbFolds)
+# DT_detector = grid_DT.fit(X_train, y_train)
+# desc_params = [DT_detector.best_params_["classifier__max_depth"]]
+# description = "Classif_" + "DT" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
+# return description, DT_detector
-def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
+def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
pipeline_DT = Pipeline([('classifier', DecisionTreeClassifier())])
param_DT = {"classifier__max_depth":np.random.randint(1, 30, 10)}
-
+ metricModule = getattr(Metrics, metric[0])
+ scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
grid_DT = GridSearchCV(pipeline_DT, param_grid=param_DT, refit=True, n_jobs=nbCores, scoring='accuracy',
cv=nbFolds)
DT_detector = grid_DT.fit(X_train, y_train)
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/KNN.py
index 60cc21c1..5e513325 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.grid_search import GridSearchCV
import numpy as np
+import Metrics
def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
nNeighbors = int(kwargs['0'])
@@ -10,20 +11,24 @@ def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
return "No desc", classifier
-def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
- pipeline_KNN = Pipeline([('classifier', KNeighborsClassifier())])
- param_KNN = {"classifier__n_neighbors": map(int, kwargs['0'])}
- grid_KNN = GridSearchCV(pipeline_KNN, param_grid=param_KNN, refit=True, n_jobs=nbCores, scoring='accuracy',
- cv=nbFolds)
- KNN_detector = grid_KNN.fit(X_train, y_train)
- desc_params = [KNN_detector.best_params_["classifier__n_neighbors"]]
- description = "Classif_" + "Lasso" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
- return description, KNN_detector
+# def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+# pipeline_KNN = Pipeline([('classifier', KNeighborsClassifier())])
+# param_KNN = {"classifier__n_neighbors": map(int, kwargs['0'])}
+# metricModule = getattr(Metrics, metric[0])
+# scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
+# grid_KNN = GridSearchCV(pipeline_KNN, param_grid=param_KNN, refit=True, n_jobs=nbCores, scoring='accuracy',
+# cv=nbFolds)
+# KNN_detector = grid_KNN.fit(X_train, y_train)
+# desc_params = [KNN_detector.best_params_["classifier__n_neighbors"]]
+# description = "Classif_" + "Lasso" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
+# return description, KNN_detector
-def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
+def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
pipeline_KNN = Pipeline([('classifier', KNeighborsClassifier())])
param_KNN = {"classifier__n_neighbors": np.random.randint(1, 30, 10)}
+ metricModule = getattr(Metrics, metric[0])
+ scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
grid_KNN = GridSearchCV(pipeline_KNN, param_grid=param_KNN, refit=True, n_jobs=nbCores, scoring='accuracy',
cv=nbFolds)
KNN_detector = grid_KNN.fit(X_train, y_train)
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
index 1b3a1f41..445fdfec 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/RandomForest.py
@@ -1,6 +1,7 @@
from sklearn.ensemble import RandomForestClassifier
from sklearn.pipeline import Pipeline
from sklearn.grid_search import GridSearchCV
+import Metrics
def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
@@ -11,41 +12,42 @@ def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
return "No desc", classifier
-def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
-
- # PipeLine with RandomForest classifier
- pipeline_rf = Pipeline([('classifier', RandomForestClassifier())])
-
- # Parameters for GridSearch: Number of Trees
- # can be extended with: oob_score, min_samples_leaf, max_features
- param_rf = {"classifier__n_estimators": map(int, kwargs['0'])}
-
- # pipeline: Gridsearch avec le pipeline comme estimator
- # param: pour obtenir le meilleur model il va essayer tous les possiblites
- # refit: pour utiliser le meilleur model apres girdsearch
- # n_jobs: Nombre de CPU (Mon ordi a des problemes avec -1 (Bug Python 2.7 sur Windows))
- # scoring: scoring...
- # cv: Nombre de K-Folds pour CV
- grid_rf = GridSearchCV(
- pipeline_rf,
- param_grid=param_rf,
- refit=True,
- n_jobs=nbCores,
- scoring='accuracy',
- cv=nbFolds,
- )
-
- rf_detector = grid_rf.fit(X_train, y_train)
-
- desc_estimators = [rf_detector.best_params_["classifier__n_estimators"]]
- description = "Classif_" + "RF" + "-" + "CV_" + str(nbFolds) + "-" + "Trees_" + str(map(str,desc_estimators))
- return description, rf_detector
-
-
-def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
+# def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+#
+# # PipeLine with RandomForest classifier
+# pipeline_rf = Pipeline([('classifier', RandomForestClassifier())])
+#
+# # Parameters for GridSearch: Number of Trees
+# # can be extended with: oob_score, min_samples_leaf, max_features
+# param_rf = {"classifier__n_estimators": map(int, kwargs['0'])}
+#
+# # pipeline: Gridsearch avec le pipeline comme estimator
+# # param: pour obtenir le meilleur model il va essayer tous les possiblites
+# # refit: pour utiliser le meilleur model apres girdsearch
+# # n_jobs: Nombre de CPU (Mon ordi a des problemes avec -1 (Bug Python 2.7 sur Windows))
+# # scoring: scoring...
+# # cv: Nombre de K-Folds pour CV
+# grid_rf = GridSearchCV(
+# pipeline_rf,
+# param_grid=param_rf,
+# refit=True,
+# n_jobs=nbCores,
+# scoring='accuracy',
+# cv=nbFolds,
+# )
+#
+# rf_detector = grid_rf.fit(X_train, y_train)
+#
+# desc_estimators = [rf_detector.best_params_["classifier__n_estimators"]]
+# description = "Classif_" + "RF" + "-" + "CV_" + str(nbFolds) + "-" + "Trees_" + str(map(str,desc_estimators))
+# return description, rf_detector
+
+
+def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
pipeline_rf = Pipeline([('classifier', RandomForestClassifier())])
param_rf = {"classifier__n_estimators": np.random.randint(1, 30, 10)}
-
+ metricModule = getattr(Metrics, metric[0])
+ scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
grid_rf = GridSearchCV(pipeline_rf,param_grid=param_rf,refit=True,n_jobs=nbCores,scoring='accuracy',cv=nbFolds)
rf_detector = grid_rf.fit(X_train, y_train)
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
index 9ad1f292..15627703 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SGD.py
@@ -2,6 +2,7 @@ from sklearn.linear_model import SGDClassifier
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.grid_search import GridSearchCV
import numpy as np
+import Metrics
def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
@@ -16,26 +17,30 @@ def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
return "No desc", classifier
-def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
- pipeline_SGD = Pipeline([('classifier', SGDClassifier())])
- param_SGD = {"classifier__loss": kwargs['1'], "classifier__penalty": kwargs['2'],
- "classifier__alpha": map(float, kwargs['0'])}
- grid_SGD = GridSearchCV(pipeline_SGD, param_grid=param_SGD, refit=True, n_jobs=nbCores, scoring='accuracy',
- cv=nbFolds)
- 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"]]
- description = "Classif_" + "Lasso" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
- return description, SGD_detector
+# def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+# pipeline_SGD = Pipeline([('classifier', SGDClassifier())])
+# param_SGD = {"classifier__loss": kwargs['1'], "classifier__penalty": kwargs['2'],
+# "classifier__alpha": map(float, kwargs['0'])}
+# metricModule = getattr(Metrics, metric[0])
+# scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
+# grid_SGD = GridSearchCV(pipeline_SGD, param_grid=param_SGD, refit=True, n_jobs=nbCores, scoring='accuracy',
+# cv=nbFolds)
+# 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"]]
+# description = "Classif_" + "Lasso" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
+# return description, SGD_detector
-def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
+def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
pipeline_SGD = Pipeline([('classifier', SGDClassifier())])
losses = ['hinge', 'log', 'modified_huber', 'squared_hinge', 'perceptron']
penalties = ["l1", "l2", "elasticnet"]
alphas = list(np.random.randint(1,10,10))+list(np.random.random_sample(10))
param_SGD = {"classifier__loss": losses, "classifier__penalty": penalties,
"classifier__alpha": alphas}
+ metricModule = getattr(Metrics, metric[0])
+ scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
grid_SGD = GridSearchCV(pipeline_SGD, param_grid=param_SGD, refit=True, n_jobs=nbCores, scoring='accuracy',
cv=nbFolds)
SGD_detector = grid_SGD.fit(X_train, y_train)
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMLinear.py
index 6f59cb29..43619432 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.grid_search import GridSearchCV
import numpy as np
+import Metrics
def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
@@ -11,22 +12,27 @@ def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
return "No desc", classifier
-def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
- pipeline_SVMLinear = Pipeline([('classifier', SVC(kernel="linear"))])
- param_SVMLinear = {"classifier__C": map(int, kwargs['0'])}
- grid_SVMLinear = GridSearchCV(pipeline_SVMLinear, param_grid=param_SVMLinear, refit=True, n_jobs=nbCores, scoring='accuracy',
- cv=nbFolds)
- SVMLinear_detector = grid_SVMLinear.fit(X_train, y_train)
- desc_params = [SVMLinear_detector.best_params_["classifier__C"]]
- description = "Classif_" + "SVC" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
- return description, SVMLinear_detector
+# def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+# pipeline_SVMLinear = Pipeline([('classifier', SVC(kernel="linear"))])
+# param_SVMLinear = {"classifier__C": map(int, kwargs['0'])}
+# metricModule = getattr(Metrics, metric[0])
+# scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
+# grid_SVMLinear = GridSearchCV(pipeline_SVMLinear, param_grid=param_SVMLinear, refit=True, n_jobs=nbCores, scoring='accuracy',
+# cv=nbFolds)
+# SVMLinear_detector = grid_SVMLinear.fit(X_train, y_train)
+# desc_params = [SVMLinear_detector.best_params_["classifier__C"]]
+# description = "Classif_" + "SVC" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
+# return description, SVMLinear_detector
-def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
+def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
pipeline_SVMLinear = Pipeline([('classifier', SVC(kernel="linear"))])
param_SVMLinear = {"classifier__C":np.random.randint(1,2000,30)}
+ metricModule = getattr(Metrics, metric[0])
+ scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
grid_SVMLinear = GridSearchCV(pipeline_SVMLinear, param_grid=param_SVMLinear, refit=True, n_jobs=nbCores, scoring='accuracy',
cv=nbFolds)
+
SVMLinear_detector = grid_SVMLinear.fit(X_train, y_train)
desc_params = [SVMLinear_detector.best_params_["classifier__C"]]
return desc_params
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
index 1c7ee599..7db4dd56 100644
--- a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
+++ b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMPoly.py
@@ -2,6 +2,7 @@ from sklearn.svm import SVC
from sklearn.pipeline import Pipeline # Pipelining in classification
from sklearn.grid_search import GridSearchCV
import numpy as np
+import Metrics
def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
@@ -12,17 +13,29 @@ def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
return "No desc", classifier
-def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
- pipeline_SVMPoly = Pipeline([('classifier', SVC(kernel="poly"))])
- param_SVMPoly= {"classifier__C": np.random.randint(1,2000,30), "classifier__degree": np.random.randint(1,10,5)}
- grid_SVMPoly = GridSearchCV(pipeline_SVMPoly, param_grid=param_SVMPoly, refit=True, n_jobs=nbCores, scoring='accuracy',
- cv=nbFolds)
- SVMPoly_detector = grid_SVMPoly.fit(X_train, y_train)
- desc_params = [SVMPoly_detector.best_params_["classifier__C"], SVMPoly_detector.best_params_["classifier__degree"]]
+# def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+# pipeline_SVMPoly = Pipeline([('classifier', SVC(kernel="poly"))])
+# param_SVMPoly= {"classifier__C": np.random.randint(1,2000,30), "classifier__degree": np.random.randint(1,10,5)}
+# metricModule = getattr(Metrics, metric[0])
+# scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
+# grid_SVMPoly = GridSearchCV(pipeline_SVMPoly, param_grid=param_SVMPoly, refit=True, n_jobs=nbCores, scoring='accuracy',
+# cv=nbFolds)
+# SVMPoly_detector = grid_SVMPoly.fit(X_train, y_train)
+# desc_params = [SVMPoly_detector.best_params_["classifier__C"], SVMPoly_detector.best_params_["classifier__degree"]]
+# return desc_params
+
+
+def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+ pipeline_SVMRBF = Pipeline([('classifier', SVC(kernel="poly"))])
+ param_SVMRBF = {"classifier__C": np.random.randint(1,2000,30)}
+ metricModule = getattr(Metrics, metric[0])
+ scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
+ grid_SVMRBF = GridSearchCV(pipeline_SVMRBF, param_grid=param_SVMRBF, refit=True, n_jobs=nbCores, scoring='accuracy',
+ cv=nbFolds)
+ SVMRBF_detector = grid_SVMRBF.fit(X_train, y_train)
+ desc_params = [SVMRBF_detector.best_params_["classifier__C"]]
return desc_params
-
-
def getConfig(config):
return "\n\t\t- SVM with C : "+config[0]+", kernel : "+config[1]
\ No newline at end of file
diff --git a/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py b/Code/MonoMutliViewClassifiers/MonoviewClassifiers/SVMRBF.py
index ae527e05..7c2e9276 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.grid_search import GridSearchCV
import numpy as np
+import Metrics
def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
@@ -11,20 +12,24 @@ def fit(DATASET, CLASS_LABELS, NB_CORES=1,**kwargs):
return "No desc", classifier
-def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
- pipeline_SVMRBF = Pipeline([('classifier', SVC(kernel="rbf"))])
- param_SVMRBF = {"classifier__C": map(int, kwargs['0'])}
- grid_SVMRBF = GridSearchCV(pipeline_SVMRBF, param_grid=param_SVMRBF, refit=True, n_jobs=nbCores, scoring='accuracy',
- cv=nbFolds)
- SVMRBF_detector = grid_SVMRBF.fit(X_train, y_train)
- desc_params = [SVMRBF_detector.best_params_["classifier__C"]]
- description = "Classif_" + "SVC" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
- return description, SVMRBF_detector
+# def fit_gridsearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
+# pipeline_SVMRBF = Pipeline([('classifier', SVC(kernel="rbf"))])
+# param_SVMRBF = {"classifier__C": map(int, kwargs['0'])}
+# metricModule = getattr(Metrics, metric[0])
+# scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
+# grid_SVMRBF = GridSearchCV(pipeline_SVMRBF, param_grid=param_SVMRBF, refit=True, n_jobs=nbCores, scoring='accuracy',
+# cv=nbFolds)
+# SVMRBF_detector = grid_SVMRBF.fit(X_train, y_train)
+# desc_params = [SVMRBF_detector.best_params_["classifier__C"]]
+# description = "Classif_" + "SVC" + "-" + "CV_" + str(nbFolds) + "-" + "-".join(map(str,desc_params))
+# return description, SVMRBF_detector
-def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, **kwargs):
+def gridSearch(X_train, y_train, nbFolds=4, nbCores=1, metric=["accuracy_score", None], **kwargs):
pipeline_SVMRBF = Pipeline([('classifier', SVC(kernel="rbf"))])
param_SVMRBF = {"classifier__C": np.random.randint(1,2000,30)}
+ metricModule = getattr(Metrics, metric[0])
+ scorer = metricModule.get_scorer(dict((index, metricConfig) for index, metricConfig in enumerate(metric[1])))
grid_SVMRBF = GridSearchCV(pipeline_SVMRBF, param_grid=param_SVMRBF, refit=True, n_jobs=nbCores, scoring='accuracy',
cv=nbFolds)
SVMRBF_detector = grid_SVMRBF.fit(X_train, y_train)
diff --git a/Code/MonoMutliViewClassifiers/Multiview/ExecMultiview.py b/Code/MonoMutliViewClassifiers/Multiview/ExecMultiview.py
index 75dceac7..450624ec 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/ExecMultiview.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/ExecMultiview.py
@@ -16,12 +16,15 @@ import time
-def ExecMultiview(DATASET, name, learningRate, nbFolds, nbCores, databaseType, path, LABELS_DICTIONARY, gridSearch=False, **kwargs):
+def ExecMultiview(DATASET, name, learningRate, nbFolds, nbCores, databaseType, path, LABELS_DICTIONARY,
+ gridSearch=False, metrics=None,**kwargs):
datasetLength = DATASET.get("Metadata").attrs["datasetLength"]
NB_VIEW = DATASET.get("Metadata").attrs["nbView"]
views = [str(DATASET.get("View"+str(viewIndex)).attrs["name"]) for viewIndex in range(NB_VIEW)]
NB_CLASS = DATASET.get("Metadata").attrs["nbClass"]
+ if not metrics:
+ metrics = ["accuracy_score" for view in range (NB_VIEW)]
CL_type = kwargs["CL_type"]
views = kwargs["views"]
@@ -78,7 +81,8 @@ def ExecMultiview(DATASET, name, learningRate, nbFolds, nbCores, databaseType, p
if gridSearch:
logging.info("Start:\t Gridsearching best settings for monoview classifiers")
- bestSettings, fusionConfig = classifierGridSearch(DATASET, classificationKWARGS, learningIndices)
+ bestSettings, fusionConfig = classifierGridSearch(DATASET, classificationKWARGS, learningIndices
+ , metrics=metrics)
classificationKWARGS["classifiersConfigs"] = bestSettings
try:
classificationKWARGS["fusionMethodConfig"] = fusionConfig
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
index 899350e0..73d82040 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Fusion.py
@@ -1,8 +1,22 @@
from Methods import *
import MonoviewClassifiers
+import numpy as np
-def gridSearch_hdf5(DATASET, classificationKWARGS):
+def makeMonoviewData_hdf5(DATASET, weights=None, usedIndices=None):
+ if not usedIndices:
+ uesdIndices = range(DATASET.get("Metadata").attrs["datasetLength"])
+ NB_VIEW = DATASET.get("Metadata").attrs["nbView"]
+ if type(weights)=="NoneType":
+ weights = np.array([1/NB_VIEW for i in range(NB_VIEW)])
+ if sum(weights)!=1:
+ weights = weights/sum(weights)
+ monoviewData = np.concatenate([weights[viewIndex]*DATASET.get("View"+str(viewIndex))[usedIndices, :]
+ for viewIndex in np.arange(NB_VIEW)], axis=1)
+ return monoviewData
+
+
+def gridSearch_hdf5(DATASET, classificationKWARGS, learningIndices, metrics=None):
fusionTypeName = classificationKWARGS["fusionType"]
fusionTypePackage = globals()[fusionTypeName+"Package"]
fusionMethodModuleName = classificationKWARGS["fusionMethod"]
@@ -12,10 +26,14 @@ def gridSearch_hdf5(DATASET, classificationKWARGS):
for classifierIndex, classifierName in enumerate(classifiersNames):
classifierModule = getattr(MonoviewClassifiers, classifierName)
classifierMethod = getattr(classifierModule, "gridSearch")
- bestSettings.append(classifierMethod(DATASET.get("View"+str(classifierIndex))[...],
- DATASET.get("labels")[...]))
+ if fusionMethodModuleName == "LateFusion":
+ bestSettings.append(classifierMethod(DATASET.get("View"+str(classifierIndex))[learningIndices],
+ DATASET.get("labels")[learningIndices], metrics=metrics[classifierIndex]))
+ else:
+ bestSettings.append(classifierMethod(makeMonoviewData_hdf5(DATASET, usedIndices=learningIndices),
+ DATASET.get("labels")[learningIndices], metrics=metrics[classifierIndex]))
classificationKWARGS["classifiersConfigs"] = bestSettings
- fusionMethodConfig = fusionMethodModule.gridSearch(DATASET, classificationKWARGS)
+ fusionMethodConfig = fusionMethodModule.gridSearch(DATASET, classificationKWARGS, learningIndices)
return bestSettings, fusionMethodConfig
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/EarlyFusion.py b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/EarlyFusion.py
index 5c0875c3..627e1bb4 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/EarlyFusion.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Fusion/Methods/EarlyFusion.py
@@ -27,50 +27,50 @@ class EarlyFusionClassifier(object):
-class WeightedLinear(EarlyFusionClassifier):
- def __init__(self, NB_CORES=1, **kwargs):
- EarlyFusionClassifier.__init__(self, kwargs['classifiersNames'], kwargs['monoviewClassifiersConfigs'],
- NB_CORES=NB_CORES)
- self.weights = np.array(map(float, kwargs['fusionMethodConfig'][0]))
-
- def fit_hdf5(self, DATASET, trainIndices=None):
- if not trainIndices:
- trainIndices = range(DATASET.get("Metadata").attrs["datasetLength"])
- self.makeMonoviewData_hdf5(DATASET, weights=self.weights, usedIndices=trainIndices)
- monoviewClassifierModule = getattr(MonoviewClassifiers, self.monoviewClassifierName)
- desc, self.monoviewClassifier = monoviewClassifierModule.fit(self.monoviewData, DATASET.get("labels")[trainIndices],
- NB_CORES=self.nbCores,
- **dict((str(configIndex),config) for configIndex,config in
- enumerate(self.monoviewClassifiersConfig)))
-
- def predict_hdf5(self, DATASET, usedIndices=None):
- if usedIndices == None:
- usedIndices = range(DATASET.get("Metadata").attrs["datasetLength"])
- if usedIndices:
- self.makeMonoviewData_hdf5(DATASET, weights=self.weights, usedIndices=usedIndices)
- predictedLabels = self.monoviewClassifier.predict(self.monoviewData)
- else:
- predictedLabels=[]
- return predictedLabels
-
- def predict_proba_hdf5(self, DATASET, usedIndices=None):
- if usedIndices == None:
- usedIndices = range(DATASET.get("Metadata").attrs["datasetLength"])
- if usedIndices:
- self.makeMonoviewData_hdf5(DATASET, weights=self.weights, usedIndices=usedIndices)
- predictedLabels = self.monoviewClassifier.predict_proba(self.monoviewData)
- else:
- predictedLabels=[]
- return predictedLabels
-
- def getConfig(self, fusionMethodConfig ,monoviewClassifiersNames, monoviewClassifiersConfigs):
- configString = "with weighted concatenation, using weights : "+", ".join(map(str, self.weights))+\
- " with monoview classifier : "
- monoviewClassifierModule = getattr(MonoviewClassifiers, monoviewClassifiersNames[0])
- configString += monoviewClassifierModule.getConfig(monoviewClassifiersConfigs[0])
- return configString
-
- def gridSearch(self, classificationKWARGS):
-
- return
+# class WeightedLinear(EarlyFusionClassifier):
+# def __init__(self, NB_CORES=1, **kwargs):
+# EarlyFusionClassifier.__init__(self, kwargs['classifiersNames'], kwargs['monoviewClassifiersConfigs'],
+# NB_CORES=NB_CORES)
+# self.weights = np.array(map(float, kwargs['fusionMethodConfig'][0]))
+#
+# def fit_hdf5(self, DATASET, trainIndices=None):
+# if not trainIndices:
+# trainIndices = range(DATASET.get("Metadata").attrs["datasetLength"])
+# self.makeMonoviewData_hdf5(DATASET, weights=self.weights, usedIndices=trainIndices)
+# monoviewClassifierModule = getattr(MonoviewClassifiers, self.monoviewClassifierName)
+# desc, self.monoviewClassifier = monoviewClassifierModule.fit(self.monoviewData, DATASET.get("labels")[trainIndices],
+# NB_CORES=self.nbCores,
+# **dict((str(configIndex),config) for configIndex,config in
+# enumerate(self.monoviewClassifiersConfig)))
+#
+# def predict_hdf5(self, DATASET, usedIndices=None):
+# if usedIndices == None:
+# usedIndices = range(DATASET.get("Metadata").attrs["datasetLength"])
+# if usedIndices:
+# self.makeMonoviewData_hdf5(DATASET, weights=self.weights, usedIndices=usedIndices)
+# predictedLabels = self.monoviewClassifier.predict(self.monoviewData)
+# else:
+# predictedLabels=[]
+# return predictedLabels
+#
+# def predict_proba_hdf5(self, DATASET, usedIndices=None):
+# if usedIndices == None:
+# usedIndices = range(DATASET.get("Metadata").attrs["datasetLength"])
+# if usedIndices:
+# self.makeMonoviewData_hdf5(DATASET, weights=self.weights, usedIndices=usedIndices)
+# predictedLabels = self.monoviewClassifier.predict_proba(self.monoviewData)
+# else:
+# predictedLabels=[]
+# return predictedLabels
+#
+# def getConfig(self, fusionMethodConfig ,monoviewClassifiersNames, monoviewClassifiersConfigs):
+# configString = "with weighted concatenation, using weights : "+", ".join(map(str, self.weights))+\
+# " with monoview classifier : "
+# monoviewClassifierModule = getattr(MonoviewClassifiers, monoviewClassifiersNames[0])
+# configString += monoviewClassifierModule.getConfig(monoviewClassifiersConfigs[0])
+# return configString
+#
+# def gridSearch(self, classificationKWARGS):
+#
+# return
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Classifiers/DecisionTree.py b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Classifiers/DecisionTree.py
index 5aa54619..97d57a57 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Classifiers/DecisionTree.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Classifiers/DecisionTree.py
@@ -31,7 +31,7 @@ def getConfig(classifierConfig):
return 'with depth ' + str(depth) + ', ' + ' sub-sampled at ' + str(subSampling) + ' '
-def gridSearch(data, labels):
+def gridSearch(data, labels, metrics="accuracy_score"):
minSubSampling = 1.0/(len(labels)/2)
bestSettings = []
bestResults = []
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Classifiers/Kover.py b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Classifiers/Kover.py
index f0e3dc15..ae1123e8 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Classifiers/Kover.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Classifiers/Kover.py
@@ -6,9 +6,4 @@ from ModifiedMulticlass import OneVsRestClassifier
# Add weights
-
-def Kover(data, labels, arg, weights,):
- isBad = False
- subSamplingRatio = arg[0]
-
- return classifier, prediction, isBad
\ No newline at end of file
+pass
diff --git a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
index c4a02a28..19dfc884 100644
--- a/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
+++ b/Code/MonoMutliViewClassifiers/Multiview/Mumbo/Mumbo.py
@@ -21,6 +21,7 @@ def computeWeights(DATASET_LENGTH, iterIndex, viewIndice, CLASS_LABELS, costMatr
for exampleIndice in range(DATASET_LENGTH)])
return weights
+
def trainWeakClassifier(classifierName, monoviewDataset, CLASS_LABELS,
DATASET_LENGTH, viewIndice, classifier_config, iterIndex, costMatrices):
weights = computeWeights(DATASET_LENGTH, iterIndex, viewIndice, CLASS_LABELS, costMatrices)
@@ -30,6 +31,7 @@ def trainWeakClassifier(classifierName, monoviewDataset, CLASS_LABELS,
logging.debug("\t\t\tView " + str(viewIndice) + " : " + str(averageAccuracy))
return classifier, classes, isBad, averageAccuracy
+
def trainWeakClassifier_hdf5(classifierName, monoviewDataset, CLASS_LABELS, DATASET_LENGTH,
viewIndice, classifier_config, viewName, iterIndex, costMatrices):
weights = computeWeights(DATASET_LENGTH, iterIndex, viewIndice, CLASS_LABELS, costMatrices)
@@ -39,22 +41,20 @@ def trainWeakClassifier_hdf5(classifierName, monoviewDataset, CLASS_LABELS, DATA
logging.debug("\t\t\tView " + str(viewIndice) + " : " + str(averageAccuracy))
return classifier, classes, isBad, averageAccuracy
-def gridSearch_hdf5(DATASET, classificationKWARGS):
+
+def gridSearch_hdf5(DATASET, classificationKWARGS, learningIndices, metrics=None):
classifiersNames = classificationKWARGS["classifiersNames"]
bestSettings = []
for classifierIndex, classifierName in enumerate(classifiersNames):
logging.debug("\tStart:\t Gridsearch for "+classifierName+" on "+DATASET.get("View"+str(classifierIndex)).attrs["name"])
classifierModule = globals()[classifierName] # Permet d'appeler une fonction avec une string
classifierMethod = getattr(classifierModule, "gridSearch")
- bestSettings.append(classifierMethod(DATASET.get("View"+str(classifierIndex))[...],
- DATASET.get("labels")[...]))
+ bestSettings.append(classifierMethod(DATASET.get("View"+str(classifierIndex))[learningIndices],
+ DATASET.get("labels")[learningIndices], metrics=metrics[classifierIndex]))
logging.debug("\tDone:\t Gridsearch for "+classifierName)
return bestSettings, None
-
-
-
class Mumbo:
def __init__(self, NB_VIEW, DATASET_LENGTH, CLASS_LABELS, NB_CORES=1,**kwargs):
--
GitLab