# In[8]:
#calling gridsearchcv
grid = utils.grid_search_cv(X_train, y_train, scoring_method="accuracy")
"""
print(grid.cv_results_)
print("============================")
print(grid.best_estimator_)
print("============================")
print(grid.best_score_)
print("============================")
print(grid.best_params_)
print("============================")
"""
# Prediction using gini
y_pred_gini = utils.prediction(X_test, grid.best_estimator_)
print("Results for gini algo")
utils.cal_accuracy(y_test, y_pred_gini)
output_filename = "%s/ML/DT/python/best_fit_graphviz_b_cancer_accuracy.txt" % q_src_dir
export_graphviz(grid.best_estimator_,
out_file=output_filename,
filled=True,
rounded=True,
special_characters=True,
feature_names=X_train.columns)
# Train using gini
clf_gini = utils.train_using_gini(X_train, y_train)
#pickle_path = "dt_gini.pkl"
#utils.save(clf_gini, pickle_path)
# In[10]:
# Train using gini
clf_gini = utils.train_using_gini(X_train, y_train)
# print(X_train[1])
export_graphviz(clf_gini,
out_file=graphviz_gini,
filled=True,
rounded=True,
special_characters=True,
feature_names=X_train.columns)
# In[11]:
# Prediction using test data and gini
y_pred_gini = utils.prediction(X_test, clf_gini)
print("Results for gini algo")
utils.cal_accuracy(y_test, y_pred_gini)
# In[12]:
# Train using entropy
clf_entropy = utils.tarin_using_entropy(X_train, y_train)
# print(clf_entropy)
utils.export_graphviz(clf_entropy, out_file=graphviz_entropy)
# In[13]:
# Prediction using entropy
y_pred_entropy = utils.prediction(X_test, clf_entropy)
print("Results for entropy algo")