import pandas as pd
import numpy as np
from pandas_profiling import ProfileReport
from sklearn.preprocessing import OneHotEncoder,LabelEncoder
from sklearn.impute import SimpleImputer
from sklearn.compose import make_column_transformer
from sklearn.preprocessing import StandardScaler

#Importing Visualization module
import matplotlib.pyplot as plt

# Model Building and Evaluation modules
from sklearn.model_selection import cross_validate
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import cross_val_score
from sklearn.pipeline import make_pipeline
from sklearn.metrics import roc_auc_score

# Importing calibration modules
from sklearn.calibration import calibration_curve, CalibratedClassifierCV

# Importing the train, Valid(Calibration) & test data
trainData = pd.read_csv("./trainNew.csv")
ValidData = pd.read_csv("./ValidData.csv")
testData = pd.read_csv("./test.csv")


labelTrain  = trainData['Status']
labelTest = testData['Status']
labelValid = ValidData['Status']

test = testData.drop(['POLICY_NO','CANCEL_CODE','Status','HIV_TEST_DATE','POLICY_HOLDER_ID'], axis = 1)
Valid = ValidData.drop(['POLICY_NO','CANCEL_CODE','Status','HIV_TEST_DATE','POLICY_HOLDER_ID'], axis = 1)
train = trainData.drop(['POLICY_NO','CANCEL_CODE','Status','HIV_TEST_DATE','POLICY_HOLDER_ID'], axis = 1)

# check missing values in train data
train.isnull().sum()

prof = ProfileReport(data)
prof.to_file(output_file='output.html')

# check column names
train.columns

# Checking data dimension'
train.shape

# Imputing missing values, Encoding Categorical Variables & Standardizing the data
column_trans = make_column_transformer(
                                        (OneHotEncoder(), ['PRODUCT_LINE_ID','SMOKING_STATUS','gender','Cover_Type']),
                                        remainder = StandardScaler()
                                       )

column_trans.fit_transform(train)

# Create a pipeline that scales the data then trains a classifier
logreg = LogisticRegression()
model_pipeline = make_pipeline(column_trans, logreg)  

# KFold/StratifiedKFold cross validation with 5 folds (the default)
# applying the classifier pipeline to the features and target data
scores = cross_val_score(model_pipeline, train, labelTrain, cv=5)
scores.mean()

# Fitting the model pipeline
model_pipeline.fit(train,labelTrain)

# Testing the model pipeline on new data/test data
predictions = model_pipeline.predict_proba(test)[:,1]
        
calib_clf = CalibratedClassifierCV(model_pipeline, method="sigmoid", cv="prefit")
calib_clf.fit(Valid, labelValid)