import pandas as pd
import numpy as np
from sklearn.linear_model import LogisticRegression

import pandas as pd
import numpy as np
import lightgbm as lgb
from sklearn.metrics import classification_report

from sklearn.model_selection import train_test_split

from sklearn.ensemble import RandomForestClassifier
from drugi_cas.modeli import classification_model

# Ucitavanje podataka

df_train = pd.read_csv(r'loan_podaci/train_ctrUa4K.csv')
df_test = pd.read_csv(r'loan_podaci/test_lAUu6dG.csv')

# Spajanje u zajednicki dataframe:
df = pd.concat([df_train, df_test])


# postavljanje indeksa
df.set_index('Loan_ID', inplace=True)


df['Self_Employed'].fillna('No', inplace=True)

def zamena_nedostajucih(x):
    agregacija = df.pivot_table(values='LoanAmount', index='Self_Employed', columns='Education', aggfunc=np.median)
    return agregacija.loc[x['Self_Employed'], x['Education']]


df['LoanAmount'].fillna(df[df['LoanAmount'].isnull()].apply(zamena_nedostajucih, axis=1), inplace=True)

# kreiranje novih kolona
df['LoanAmount_log'] = np.log(df['LoanAmount'])
df['TotalIncome'] = df['ApplicantIncome'] + df['CoapplicantIncome']
df['TotalIncome_log'] = np.log(df['TotalIncome'])

df['Gender'].fillna(df['Gender'].mode()[0], inplace=True)
df['Married'].fillna(df['Married'].mode()[0], inplace=True)
df['Dependents'].fillna(df['Dependents'].mode()[0], inplace=True)
df['Loan_Amount_Term'].fillna(df['Loan_Amount_Term'].mode()[0], inplace=True)
df['Credit_History'].fillna(df['Credit_History'].mode()[0], inplace=True)

# enkodiranje promennjivih kategorickog tipa
for kolona in df.columns:
    if df[kolona]._is_numeric_mixed_type == False:
        df[kolona] = df[kolona].astype('category').cat.codes.astype('category')

x_train = df.iloc[:df_train.shape[0]].drop('Loan_Status', axis=1)
y_train = df.iloc[:df_train.shape[0]].Loan_Status  # ili df_train['Loan_Status']

x_test = df.iloc[df_train.shape[0]:].drop('Loan_Status', axis=1)


model = LogisticRegression(solver='lbfgs')
predictor_var = ['Credit_History']
rezultat = classification_model(model, x_train, predictor_var, y_train, x_test)

prag = 0.5
x_test['Loan_Status'] = np.where(rezultat > prag, 'Y', 'N')
x_test['Loan_Status'].reset_index().to_csv('loan_podaci/primer_izvoza_logisticka.csv', index=False)