From d3b6fe13a6bcb2a11d1ddd9e91191ea89ba60bb5 Mon Sep 17 00:00:00 2001
From: Tanushree Tunstall <tanu@tunstall.in>
Date: Fri, 18 Mar 2022 17:51:38 +0000
Subject: [PATCH] added grid_search_vs_base_estimator.py to compare results
 from baseestimator and gridsearch manual

---
 grid_search_vs_base_estimator.py | 82 ++++++++++++++++++++++++++++++++
 1 file changed, 82 insertions(+)
 create mode 100644 grid_search_vs_base_estimator.py

diff --git a/grid_search_vs_base_estimator.py b/grid_search_vs_base_estimator.py
new file mode 100644
index 0000000..ae9cff5
--- /dev/null
+++ b/grid_search_vs_base_estimator.py
@@ -0,0 +1,82 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+Created on Wed Mar 16 16:55:06 2022
+
+@author: tanu
+"""
+# https://stackoverflow.com/questions/57248072/gridsearchcv-gives-different-result
+#%% variables
+mcc_score_fn = {'mcc': make_scorer(matthews_corrcoef)}
+
+scoring_refit = {'scoring': 'recall'
+                 ,'refit': 'recall'}
+
+scoring_refit = {'scoring': mcc_score_fn
+                 , 'refit': 'mcc'}
+
+scoring_refit = {'scoring': 'jaccard'
+                 ,'refit': 'jaccard'}
+
+njobs = {'n_jobs': 10} # my desktop has 12 cores
+skf_cv = StratifiedKFold(n_splits=10,shuffle  = True)
+cv = {'cv': 10}
+
+#%% GridSearchCV
+gs_dt = GridSearchCV(estimator=DecisionTreeClassifier(**rs
+                                                      #,class_weight = {1:10, 0:1}
+                                                      ),
+                  param_grid=[{'max_depth': [ 2,  4,  6, 8, 10]
+                               , 'criterion':['gini','entropy']
+                               , "max_features":["auto", None]
+                               , "max_leaf_nodes":[10,20,30,40]}]
+                  #, **cv
+                  , cv = skf_cv
+                  , **scoring_refit)
+                  #, scoring = mcc_score_fn, refit = 'mcc')
+
+#gs_dt.fit(X_train, y_train)
+#gs_dt_fit = gs_dt.fit(X_train, y_train)
+gs_dt.fit(X, y)
+gs_dt_fit = gs_dt.fit(X, y)
+gs_dt_fit_res  = gs_dt_fit.cv_results_
+
+print('Best model:\n', gs_dt.best_params_)
+print('Best models score:\n', gs_dt.best_score_)
+
+best_model = gs_dt.best_params_
+best_model
+best_model_score =  gs_dt.best_score_
+best_model_score
+#%% Check the scores:
+#  https://stackoverflow.com/questions/44947574/what-is-the-meaning-of-mean-test-score-in-cv-result
+#print([(len(train), len(test)) for train, test in skf_cv.split(X, y)])
+gs_dt_fit.cv_results_
+#%%BaseEstimator:Recall
+pipeline = Pipeline([
+    ('pre', MinMaxScaler()),
+    ('clf', ClfSwitcher()),
+])
+
+
+parameters = [
+    {
+        'clf__estimator': [DecisionTreeClassifier(**rs)]
+        , 'clf__estimator__max_depth': [ 2,  4,  6, 8, 10]
+        , 'clf__estimator__criterion':['gini','entropy']
+        , 'clf__estimator__max_features':["auto", None]
+        , 'clf__estimator__max_leaf_nodes':[10,20,30,40]
+    }
+]
+
+gscv = GridSearchCV(pipeline
+                    , parameters
+                    , cv=skf_cv
+                    , **njobs
+                    , **scoring_refit
+                    , return_train_score=False
+                    , verbose=3)
+
+gscv.fit(X, y)
+print('Best model:\n', gscv.best_params_)
+print('Best models score:\n', gscv.best_score_)
\ No newline at end of file