def test_get_item_by_id(self):
firebase = get_storage()._firebase
for resource in self._resources:
data = firebase.get('/data', None).get(resource)
not_none_element_id = 0
for element in data:
if not (element is None):
not_none_element_id = element.get('id')
break
result = get_storage().get_item_by_id(resource,
not_none_element_id)
self.assertIsNotNone(result.get('id'))
def test_connection(self):
firebase = get_storage()._firebase
versions = firebase.get('/versions', None)
data = firebase.get('/data', None)
for resource in self._resources:
self.assertIsNotNone(versions.get(resource))
self.assertIsNotNone(data.get(resource))
def __init__(self, resource: str):
super().__init__("sync_cbc_%s" % (resource))
self._ps = get_ps()
self._ss = get_storage()
self._resource = resource
self._iso_lang = get_config()['prestashop']['mainLanguage']
self._current_lang = self.get_language(self._iso_lang)
def test_get_items_older_version(self):
for resource in self._resources:
results = get_storage().get_items_older_version(resource)
latest_version = get_storage().latest_version(resource)
for result in results:
self.assertNotEqual(result.get('version'), latest_version)
def test_get_items(self):
for resource in self._resources:
results = get_storage().get_items(resource)
self.assertTrue(type(results), list)
for result in results:
self.assertIsNotNone(result)
def test_latest_version(self):
for resource in self._resources:
self.assertEqual(type(get_storage().latest_version(resource)), int)
error_list = cross_val_score(model,
X_train,
y_train,
cv=3,
scoring='neg_mean_squared_error')
return error_list.mean()
# #### チューニング開始
study = optuna.create_study(direction='maximize',
pruner=optuna.pruners.MedianPruner(),
study_name='sample',
storage=get_storage(),
load_if_exists=True)
study.optimize(objective, n_trials=50)
# +
# study = optuna.load_study(study_name='sample', storage=get_storage())
# -
# #### デフォルトパラメータとチューニングしたパラメータの比較
default_model = RandomForestRegressor(random_state=1234)
default_model.fit(X_train, y_train)
default_predict = default_model.predict(X_test)
default_score = mean_squared_error(y_test, default_predict)
# +