class _RidgeClassifierCVImpl:
def __init__(self, **hyperparams):
self._hyperparams = hyperparams
self._wrapped_model = Op(**self._hyperparams)
def fit(self, X, y=None):
if y is not None:
self._wrapped_model.fit(X, y)
else:
self._wrapped_model.fit(X)
return self
def predict(self, X):
return self._wrapped_model.predict(X)
def decision_function(self, X):
return self._wrapped_model.decision_function(X)
def build_matrix_df_old(pickle_filename):
flickr_results = pickle.load(open(pickle_filename))
combined_scores, all_reqd_tags = curate_imagescores(flickr_results)
matrix, outcome, location_dict = build_all_rows(combined_scores,
all_reqd_tags)
data = {
'matrix': matrix,
'outcome': outcome,
'rows': all_reqd_tags,
'location_dict': location_dict
}
pickle.dump(data, open(os.path.join('./cache', 'flickr_data.pickle'),
'w+'))
print(matrix.wildlife.tolist())
dt = DecisionTreeClassifier(random_state=0).fit(matrix, outcome)
pickle.dump(
dt,
open(os.path.join('./cache', 'flickr_model_decisiontree.pickle'),
'w+'))
print('Feature Importance', dt.feature_importances_)
predict = dt.predict_proba(matrix)
print('Probability', predict)
act_predict = dt.predict(matrix)
print(act_predict.tolist())
score = dt.score(matrix, outcome)
print('Shape', score)
rc = RidgeClassifierCV(cv=3, normalize=True).fit(matrix, outcome)
pickle.dump(
rc,
open(os.path.join('./cache', 'flickr_model_ridgeclasscv.pickle'),
'w+'))
rc_df = rc.decision_function(matrix)
for ind_row in rc_df:
print(list(ind_row))
#print(rc_df)
raw_input('Done With Creating Model')
Y_lr = lr.predict(X_test) print(accuracy_score(Y_test, Y_lr)) # In[14]: svc = SVC(C=1.0, kernel='rbf') svc.fit(X_train, Y_train) Y_SVC = svc.predict(X_test) accuracy_score(Y_test, Y_SVC) # In[15]: rc = RidgeClassifierCV(alphas=(0.1, 1.0, 10.0)) rc.fit(X_train, Y_train) Y_rc = rc.predict(X_test) rc.decision_function(X_test) # In[16]: tr = DecisionTreeClassifier(criterion='gini') tr.fit(X_train, Y_train) Y_tr = tr.predict(X_test) accuracy_score(Y_test, Y_tr) # In[17]: rf = RandomForestClassifier(n_estimators=10, criterion='gini') rf.fit(X_train, Y_train) Y_rf = rf.predict(X_test) accuracy_score(Y_test, Y_rf)
