SVR

from sklearn.svm import SVR
from sklearn.preprocessing import MinMaxScaler
import pandas as pd 
from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score
import math
import model_evaluation
import time
"""
Parameters: 
- kernel: Specifies the kernel type to be used in the algorithm. 
  kernel = {‘linear’, ‘poly’, ‘rbf’, ‘sigmoid’, ‘precomputed’} 
- X_train: Training feature data.
- y_train: Training target data.
- X_test: Test feature data.
"""
def support_vector_regression(kernel, X_train, y_train, X_test):    
    # Initialise an SVR model with given kernel
    svr_model = SVR(kernel=kernel)
    # train and predict model
    svr_model.fit(X_train, y_train)
    svr_predictions = svr_model.predict(X_test)
    # Return predictions
    return svr_predictions


def main():
    train_set = pd.read_csv("train.csv")
    test_set = pd.read_csv("test.csv")
    features = ['OCC_YEAR','OCC_MONTH','PREMISES','HOOD','LONG','LAT','WEIGHT','NSI']
    target = 'TARGET'
    X_train = train_set[features].values
    y_train = train_set[target].values
    X_test = test_set[features].values
    y_test = test_set[target].values
    scaler = MinMaxScaler()
    X_train_scaled = scaler.fit_transform(X_train)
    X_test_scaled = scaler.transform(X_test)
    kernel = 'linear'  #Linear kernel is chosen due to its lowest computation time among all kernels
    print("Training SVR")
    start=time.time()
    y_pred = support_vector_regression(kernel, X_train_scaled, y_train, X_test_scaled)
    end=time.time() 
    print("Training Completed")
    # print the evaluation metrics using the print_evaluation in model_evaluation
    print("Training Time",end-start,"seconds")
    model_evaluation.print_evaluation(y_test, y_pred)
    print("END")
        
if __name__ == "__main__":
    main()