import lightgbm as lgb
import prepare

import xgboost as xgb
from xgboost import plot_importance
from matplotlib import pyplot as plt
from lightgbm import LGBMClassifier
from catboost import CatBoostClassifier
from sklearn.model_selection import train_test_split,cross_val_score,StratifiedKFold,GridSearchCV
from sklearn.metrics import roc_auc_score
from xgboost.sklearn import XGBClassifier

from sklearn.metrics import confusion_matrix, recall_score, roc_auc_score , accuracy_score

import prepare
import pandas as pd
import time
import datetime


def lgb_722():
    x, val_x, y, val_y, testA, sub = prepare.read()
    # print(Y.head())
    cate_features = ['grade', 'subGrade', 'employmentTitle', 'homeOwnership', 'verificationStatus', 'purpose',
                     'postCode',
                     'regionCode', 'applicationType', 'initialListStatus', 'title', 'policyCode']

    train = lgb.Dataset(data=x, label=y)
    val = lgb.Dataset(data=val_x, label=val_y)

    # valdata = dataset.create_valid()

    param = {'num_leaves': 31, 'num_trees': 100, 'objective': 'binary'}
    param['metric'] = ['auc', 'binary_logloss']

    num_round = 500
    bst = lgb.train(param, train, num_round, valid_sets=[val])

    r = bst.predict(testA)

    print(r.shape, r)

    sub['isDefault'] = r
    sub.to_csv('lightgbm.csv', index=False)




def xgb_553():
    X_train = "data/prepare/xtrain.csv"
    X_test = "data/prepare/xtest.csv"
    y_train = "data/prepare/ytrain.csv"
    y_test = "data/prepare/ytest.csv"
    testA = "data/prepare/testA.csv"

    X_train, X_test, y_train, y_test, testA = prepare.prepare_csv(X_train, X_test, y_train, y_test, testA)
    sub = pd.read_csv("data/sample_submit.csv")

    i = 5
    m = XGBClassifier(max_depth=5, n_estimators=490, eval_metric=['logloss', 'auc', 'error'])
    m.fit(X_train, y_train)
    p = m.predict(X_test)
    s = m.score(X_test, y_test)
    print(f"max_depth:{i},score", s)
    print(f"max_depth:{i},混淆矩阵", confusion_matrix(y_test, p, labels=[1, 0]))
    print(f"max_depth:{i},召回率", recall_score(y_test, p))
    print(f"max_depth:{i},auc", roc_auc_score(y_test, m.predict_proba(X_test)[:, 1]))

    r = m.predict(testA)
    sub['isDefault'] = r
    sub.to_csv('xgb-02-2.csv', index=False)



def xgb_728():
    X_train = "data/prepare/xtrain.csv"
    X_test = "data/prepare/xtest.csv"
    y_train = "data/prepare/ytrain.csv"
    y_test = "data/prepare/ytest.csv"
    testA = "data/prepare/testA.csv"

    X_train, X_test, y_train, y_test, testA = prepare.prepare_csv(X_train, X_test, y_train, y_test, testA)
    sub = pd.read_csv("data/sample_submit.csv")

    i = 5
    m = XGBClassifier(max_depth=5,n_estimators=490,eval_metric=['logloss','auc','error'])
    m.fit(X_train,y_train)
    p=m.predict(X_test)
    s=m.score(X_test,y_test)
    print(f"max_depth:{i},score",s)
    print(f"max_depth:{i},混淆矩阵",confusion_matrix(y_test,p,labels=[1,0]))
    print(f"max_depth:{i},召回率",recall_score(y_test,p))
    print(f"max_depth:{i},auc",roc_auc_score(y_test,m.predict_proba(X_test)[:,1]))

    r=m.predict_proba(testA)[:,1]
    print(r.shape,r)
    sub['isDefault'] = r
    sub.to_csv('xgb-03-1.csv', index=False)


if __name__ == '__main__':
    lgb_722()