classifying.py

import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC
from sklearn.naive_bayes import GaussianNB, MultinomialNB
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, confusion_matrix
from sklearn.model_selection import GridSearchCV
from sklearn.linear_model import LogisticRegression
from load_test_data import load

def classify(vectors, labels, train_text, task, type="DT"):
    # Random Splitting With Ratio 3 : 1
    train_vectors, test_vectors, train_labels, test_labels = train_test_split(vectors, labels, test_size=0.333)
    test_vectors, test_labels = load(train_text, task)

    
    # Initialize Model
    classifier = None
    if(type=="MNB"):
        classifier = MultinomialNB(alpha=0.7)
        classifier.fit(train_vectors, train_labels)
    elif(type=="KNN"):
        classifier = KNeighborsClassifier(n_jobs=4)
        params = {'n_neighbors': [3,5,7,9], 'weights':['uniform', 'distance']}
        classifier = GridSearchCV(classifier, params, cv=3, n_jobs=4)
        classifier.fit(train_vectors, train_labels)
        classifier = classifier.best_estimator_
    elif(type=="SVM"):
        classifier = SVC()
        classifier = GridSearchCV(classifier, {'C':[0.001, 0.01, 0.1, 1, 10]}, cv=3, n_jobs=4)
        classifier.fit(train_vectors, train_labels)
        classifier = classifier.best_estimator_
    elif(type=="DT"):
        classifier = DecisionTreeClassifier(max_depth=800, min_samples_split=5)
        params = {'criterion':['gini','entropy']}
        classifier = GridSearchCV(classifier, params, cv=3, n_jobs=4)
        classifier.fit(train_vectors, train_labels)
        classifier = classifier.best_estimator_
    elif(type=="RF"):
        classifier = RandomForestClassifier(max_depth=800, min_samples_split=5)
        params = {'n_estimators': [n for n in range(50,200,50)], 'criterion':['gini','entropy'], }
        classifier = GridSearchCV(classifier, params, cv=3, n_jobs=4)
        classifier.fit(train_vectors, train_labels)
        classifier = classifier.best_estimator_
    elif(type=="LR"):
        classifier = LogisticRegression(multi_class='auto', solver='newton-cg',)
        classifier = GridSearchCV(classifier, {"C":np.logspace(-3,3,7), "penalty":["l2"]}, cv=3, n_jobs=4)
        classifier.fit(train_vectors, train_labels)
        classifier = classifier.best_estimator_
    else:
        print("Wrong Classifier Type!")
        return

    accuracy = accuracy_score(train_labels, classifier.predict(train_vectors))
    print("Training Accuracy:", accuracy)
    test_predictions = classifier.predict(test_vectors)
    accuracy = accuracy_score(test_labels, test_predictions)
    print("Test Accuracy:", accuracy)
    print("Confusion Matrix:", )
    print(confusion_matrix(test_labels, test_predictions))