def go_by_category(category):
input = TrainingFactory.build_sparse_matrix_target(limit=10000)
targets = TrainingFactory.build_target_vector_by_category(category,limit=10000)
input_train, input_test, target_train, target_test = train_test_split(input, targets, test_size=0.1)
classif = SVC(kernel='rbf', tol=0.001, probability=True)
classif.fit(input_train, target_train)
output_targets = classif.predict(input_test)
print output_targets
print target_test
print
def go_by_category(category):
input = TrainingFactory.build_sparse_matrix_target(limit=10000)
targets = TrainingFactory.build_target_vector_by_category(category,
limit=10000)
input_train, input_test, target_train, target_test = train_test_split(
input, targets, test_size=0.1)
classif = SVC(kernel='rbf', tol=0.001, probability=True)
classif.fit(input_train, target_train)
output_targets = classif.predict(input_test)
print output_targets
print target_test
print
def go_by_category_2(category):
input, targets, scaler = TrainingFactory.get_training_data_by_category(category,10000)
input_train, input_test, target_train, target_test = train_test_split(input, targets, test_size=0.1)
test_data_sparse = TestingFactory.get_test_data(limit=1000)
test_data_scaled = scaler.transform(test_data_sparse)
test_data = csr_matrix(test_data_scaled)
classif = SVC(kernel='rbf',C=0.1, tol=0.001, probability=True)
classif.fit(input_train, target_train)
output_targets_proba = classif.predict_proba(input_test)
outputs_predicted_proba = [item[1] for item in output_targets_proba]
output_targets = classif.predict(input_test)
# print output_targets.tolist()
# print outputs_predicted_proba
# print target_test
print log_loss(target_test, output_targets)
accuracy = accuracy_score(target_test, output_targets)
print accuracy
print confusion_matrix(target_test, output_targets)
testing_output = classif.predict_proba(test_data)
testing_output_proba = [item[1] for item in testing_output]
print testing_output_proba
return accuracy, output_targets, testing_output_proba
def go():
input = TrainingFactory.build_sparse_matrix_input(limit=10000)
targets = TrainingFactory.build_sparse_matrix_target(limit=10000)
input_train, input_test, target_train, target_test = train_test_split(input, targets, test_size=0.1)
classif = OneVsRestClassifier(SVC(kernel='rbf', tol=0.001, probability=True))
classif.fit(input_train, target_train)
output_targets = classif.predict_proba(input_test)
print ClassifierFactory.output_function(output_targets)
print ClassifierFactory.output_function(target_test.todense())
print log_loss(target_test, output_targets)
print
def go_by_category_2(category):
input, targets, scaler = TrainingFactory.get_training_data_by_category(
category, 10000)
input_train, input_test, target_train, target_test = train_test_split(
input, targets, test_size=0.1)
test_data_sparse = TestingFactory.get_test_data(limit=1000)
test_data_scaled = scaler.transform(test_data_sparse)
test_data = csr_matrix(test_data_scaled)
classif = SVC(kernel='rbf', C=0.1, tol=0.001, probability=True)
classif.fit(input_train, target_train)
output_targets_proba = classif.predict_proba(input_test)
outputs_predicted_proba = [item[1] for item in output_targets_proba]
output_targets = classif.predict(input_test)
# print output_targets.tolist()
# print outputs_predicted_proba
# print target_test
print log_loss(target_test, output_targets)
accuracy = accuracy_score(target_test, output_targets)
print accuracy
print confusion_matrix(target_test, output_targets)
testing_output = classif.predict_proba(test_data)
testing_output_proba = [item[1] for item in testing_output]
print testing_output_proba
return accuracy, output_targets, testing_output_proba
def go():
input = TrainingFactory.build_sparse_matrix_input(limit=10000)
targets = TrainingFactory.build_sparse_matrix_target(limit=10000)
input_train, input_test, target_train, target_test = train_test_split(
input, targets, test_size=0.1)
classif = OneVsRestClassifier(
SVC(kernel='rbf', tol=0.001, probability=True))
classif.fit(input_train, target_train)
output_targets = classif.predict_proba(input_test)
print ClassifierFactory.output_function(output_targets)
print ClassifierFactory.output_function(target_test.todense())
print log_loss(target_test, output_targets)
print
def go_by_category_2(category,num):
#print category
#return
input, targets, scaler = TrainingFactory.get_training_data_by_category(category,10000)
# Split arrays into random train and test subsets
input_train, input_test, target_train, target_test = train_test_split(input, targets, test_size=0.1)
test_data_sparse = TestingFactory.get_test_data(limit=10000)
test_data_scaled = scaler.transform(test_data_sparse)
test_data = csr_matrix(test_data_scaled)
if(num==1):
classif = SVC(kernel='rbf',C=0.1, tol=0.001, probability=True)
elif(num==0):
classif= rfc(n_estimators=500, oob_score=True)
elif(num==2):
classif = MultinomialNB()
elif(num==3):
classif = tree.DecisionTreeClassifier(max_depth=1000)
else:
classif = SVC(kernel='linear',C=0.1, tol=0.001, probability=True)
#classif.fit(input_train, target_train)
classif.fit(input_train, target_train)
#output_targets_proba = classif.predict_proba(input_test)
#outputs_predicted_proba = [item[1] for item in output_targets_proba]
output_targets = classif.predict(input_test)
# print output_targets.tolist()
# print outputs_predicted_proba
# print target_test
print "log loss: ",log_loss(target_test, output_targets)
accuracy = accuracy_score(target_test, output_targets)
print "accuracy: ",accuracy
cm= confusion_matrix(target_test, output_targets)
print "Confusion matrix :",cm,"\n"
#print "prediction score"
#print precision_score(target_test,output_targets,average="macro")
testing_output = classif.predict_proba(test_data)
testing_output_proba = [item[1] for item in testing_output]
###
return accuracy, output_targets, testing_output_proba
