#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Sat May 21 02:52:36 2022

@author: tanu
"""
# https://scikit-learn.org/stable/modules/generated/sklearn.pipeline.Pipeline.html
import pandas as pd
from sklearn.pipeline import Pipeline
from sklearn.datasets import make_classification
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import GridSearchCV
from sklearn.neighbors import KNeighborsClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.feature_selection import SelectKBest, mutual_info_classif
#pd.options.plotting.backend = "plotly"
X_eg, y_eg = make_classification(n_samples=1000,
                           n_features=30,
                           n_informative=5,
                           n_redundant=5,
                           n_classes=2,
                           random_state=123)

pipe = Pipeline([('scaler', StandardScaler()),
                 ('selector', SelectKBest(mutual_info_classif, k=9)),
                 
                 ('classifier', LogisticRegression())])

search_space = [{'selector__k': [5, 6, 7, 10]},
                {'classifier': [LogisticRegression()],
                 'classifier__C': [0.01,1.0],
                 'classifier__solver': ['saga', 'lbfgs']},
                {'classifier': [RandomForestClassifier(n_estimators=100)],
                 'classifier__max_depth': [5, 10, None]},
                {'classifier': [KNeighborsClassifier()],
                 'classifier__n_neighbors': [3, 7, 11],
                 'classifier__weights': ['uniform', 'distance']}]



clf = GridSearchCV(pipe, search_space, cv=10, verbose=0)

clf2 = clf.fit(X_eg, y_eg)
clf2._check_feature_names
clf2.best_estimator_.named_steps['selector'].n_features_in_

clf2.best_estimator_ #n of best features
clf2.best_params_ 
clf2.best_estimator_.get_params
clf2.get_feature_names(


clf3 = clf2.best_estimator_ # 
clf3._final_estimator
clf3._final_estimator.C
clf3._final_estimator.solver


fs_bmod = clf2.best_estimator_
print('\nbest model with feature selection:', fs_bmod)

#########################################################
# my data

pipe = Pipeline([
    ('pre', MinMaxScaler())
    ('selector', RFECV(LogisticRegression(**rs), cv = skf_cv, scoring = 'matthews_corrcoef'))
    , ('classifier',  LogisticRegression(**rs))])

search_space = [{'selector__min_features_to_select': [1,2]},
                {'classifier': [LogisticRegression()],
                 #'classifier__C': np.logspace(0, 4, 10),
                 'classifier__C': [2, 2.8],
                 'classifier__max_iter': [100],
                 'classifier__penalty': ['l1', 'l2'],
                 'classifier__solver': ['saga']
                 }] #,
                #{'classifier': [RandomForestClassifier(n_estimators=100)],
                # 'classifier__max_depth': [5, 10, None]},
                #{'classifier': [KNeighborsClassifier()],
                # 'classifier__n_neighbors': [3, 7, 11],
                # 'classifier__weights': ['uniform', 'distance']
                #}]

clf = GridSearchCV(pipe, search_space, cv=skf_cv, scoring = mcc_score_fn, refit = 'mcc',  verbose=0)

clf.fit(X, y)
clf.best_params_
clf.best_score_

tp = clf.predict(X_bts)
print('\nMCC on Blind test:'     , round(matthews_corrcoef(y_bts, tp),2))
print('\nAccuracy on Blind test:', round(accuracy_score(y_bts, tp),2))