def run_all_classifiers(stop_words, use_pipeline, dynamic_datasets_path):
data = readDatasets.read_dataset(dynamic_datasets_path)
train_data = data[0]
test_data = data[1]
print('Running the classifiers...\n')
nbScores = nbClassifier.nb_classifier(stop_words, train_data, test_data,
use_pipeline)
rfScores = rfClassifier.rf_classifier(stop_words, train_data, test_data,
use_pipeline)
svmScores = svmClassifier.svm_classifier(stop_words, train_data, test_data,
use_pipeline)
# knnScores = knnClassifier.knn_classifier(stop_words, train_data, test_data)
mymethodScores = MyMethodClassifier.my_method_classifier(
stop_words, train_data, test_data, dynamic_datasets_path)
'{:06.2f}'.format(3.141592653589793)
# Open an outputCsvFile and write the scores which we will receive from the classifiers.
print('Writing classifiers\' scores to the outputCsvFile...\n')
location_10_fold = os.path.join(dynamic_datasets_path, 'Resources',
'datasets', 'EvaluationMetric_10fold.csv')
with open(location_10_fold, mode='w', encoding="utf8") as csvfile:
csvWriter = csv.writer(csvfile,
delimiter='\t',
quotechar='|',
quoting=csv.QUOTE_MINIMAL)
# Write the Headers (first row & column).
csvWriter.writerow(['Statistic_Measure'] + ['Naive_Bayes'] +
['Random_Forest'] + ['SVM'] + ['KNN'] +
['My_Method'])
# Write the scores.
csvWriter.writerow(['Accuracy'] + ['{:.3}'.format(nbScores[0])] +
['{:.3}'.format(rfScores[0])] +
['{:.3}'.format(svmScores[0])] + ['knn'] +
['{:.3}'.format(mymethodScores[0])])
# + [knnScores[0]] + [mymethodScores[0]])
csvWriter.writerow(['Precision'] + ['{:.3}'.format(nbScores[1])] +
['{:.3}'.format(rfScores[1])] +
['{:.3}'.format(svmScores[1])] + ['knn'] +
['{:.3}'.format(mymethodScores[1])])
# + [knnScores[1]]] + [mymethodScores[1]])
csvWriter.writerow(['Recall'] + ['{:.3}'.format(nbScores[2])] +
['{:.3}'.format(rfScores[2])] +
['{:.3}'.format(svmScores[2])] + ['knn'] +
['{:.3}'.format(mymethodScores[2])])
# + [knnScores[2]] + [mymethodScores[2]])
csvWriter.writerow(['F-Measure'] + ['{:.3}'.format(nbScores[3])] +
['{:.3}'.format(rfScores[3])] +
['{:.3}'.format(svmScores[3])] + ['knn'] +
['{:.3}'.format(mymethodScores[3])])
# + [knnScores[3]]] + [mymethodScores[3]])
print('Finished writing to the outputCsvFile!')
clf.fit(vectorTrain, train_data['Category'])
y_pred = clf.predict(vectorTest)
# print "Train Accuracy :: ", accuracy_score(train_data['Category'], clf.predict(vectorTrain))
# print "Test Accuracy :: ", accuracy_score(train_data['Category'], y_pred)
#y_pred = cross_val_predict(clf, X=vectorTrain, y=vectorTest, cv=10, n_jobs=multiprocessing.cpu_count())
writePredictionsToCsv.write_predictions_to_csv(y_pred, test_data,
dynamic_datasets_path)
# Best GridSearch params
# print clf.best_params_
print("Elapsed time of successional-run: ",
time.time() - start_time_successional)
print('MyMethodClassifier finished!\n')
return scores
# Run myMethodClassifier directly:
if __name__ == '__main__':
dynamic_datasets_path = '..'
data = readDatasets.read_dataset(dynamic_datasets_path)
trainData = data[0]
testData = data[1]
my_method_classifier(stopWords.get_stop_words(), trainData, testData,
dynamic_datasets_path)
exit()
# GridSearch
# parameters = {'n_estimators': [130, 110, 100, 80, 50, 30, 20, 10]}
# svr = RandomForestClassifier()
# clf = GridSearchCV(svr, parameters)
# clf.fit(vectorTrain, train_y)
# y_pred = clf.predict(vectorTest)
#
# print "Train Accuracy :: ", accuracy_score(train_y, clf.predict(vectorTrain))
# print "Test Accuracy :: ", accuracy_score(test_y, y_pred)
# Best GridSearch params
# print clf.best_params_
print("Elapsed time of successional-run: ",
time.time() - start_time_successional)
print('rfClassifier finished!\n')
return scores
# Run rfClassifier directly:
if __name__ == '__main__':
data = readDatasets.read_dataset()
trainData = data[0]
testData = data[1]
usePipeline = False
rf_classifier(stopWords.get_stop_words(), trainData, testData, usePipeline)