import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.feature_selection import VarianceThreshold
from sklearn.ensemble import RandomForestClassifier,RandomForestRegressor
from sklearn.metrics import roc_auc_score


from mlxtend.feature_selection import SequentialFeatureSelector

data = pd.read_csv("truck.csv")
print(data.shape)

output = data.iloc[:,402]
print(output)

input = data.drop(labels=['Send_Date','PARTITIONING','VEHICLE_ID','All_Fault_in_3_months'],axis=1)

print(input)




train_features,test_features,train_labels,test_labels = train_test_split(input,output,test_size=0.2,random_state=41)

correlated_features = set()
correlated_matrix = input.corr()
for i in range(len(correlated_matrix.columns)):
    for j in range(i):
        if abs(correlated_matrix.iloc[i,j] > 0.8):
            colname = correlated_matrix.columns[i]
            correlated_features.add(colname)


train_features.drop(labels=correlated_features, axis=1, inplace=True)
test_features.drop(labels=correlated_features, axis=1, inplace=True)

print(train_features.shape)
print(test_features.shape)

feature_Selector = SequentialFeatureSelector(RandomForestClassifier(n_jobs=-1),k_features=50,forward=True,verbose=2,scoring='roc_auc',cv=4)

features = feature_Selector.fit(np.array(train_features.fillna(0)),train_labels)
filtered_features= train_features.columns[list(features.k_feature_idx_)]
print(filtered_features)


#to see how it works

classifier = RandomForestClassifier(n_estimators=100,random_state=41,max_depth=6)
classifier.fit(train_features[filtered_features].fillna(0),train_labels)

train_prediction = classifier.predict_proba(train_features[filtered_features].fillna(0))
print('Accuracy on training set: {}'.format(roc_auc_score(train_labels, train_prediction[:,1])))

test_prediction = classifier.predict_proba(test_features[filtered_features].fillna(0))
print('Accuracy on test set: {}'.format(roc_auc_score(test_labels, test_prediction[:,1])))