def testLoadModel(self):
models = load_linear_models_from_avro(self.model_file, self.feature_file)
for model, expected in zip(models, self.expected_models):
self.assertAllEqual(model, expected)
short_models = load_linear_models_from_avro(self.model_file, self.short_feature_file)
for i in range(len(short_models)):
self.assertAllEqual(short_models[i], self.expected_short_models[i])
def _check_model(self, coefficients, model_dir, feature_file):
"""
Check if model coefficients are as expected
:param coefficients: Expected coefficients
:param model_dir: Model directory
:return: None
"""
model_file = os.path.join(model_dir, "part-00000.avro")
model = load_linear_models_from_avro(model_file, feature_file)[0]
self.assertAllClose(coefficients, model, msg='models mismatch')
def _load_model(self, catch_exception=False):
""" Load model from avro file. """
model = None
logging("Loading model from {}".format(self.checkpoint_path))
model_exist = self.checkpoint_path and tf1.io.gfile.exists(self.checkpoint_path)
if model_exist:
model_file = tf1.io.gfile.glob("{}/*.avro".format(self.checkpoint_path))
if len(model_file) == 1:
model = load_linear_models_from_avro(model_file[0], self.feature_file)[0]
elif not catch_exception:
raise ValueError("Load model failed, no model file or multiple model"
" files found in the model diretory {}".format(self.checkpoint))
elif not catch_exception:
raise FileNotFoundError("checkpoint path {} doesn't exist".format(self.checkpoint_path))
if self.feature_bag_name is None and model is not None:
# intercept only model, add a dummy weight.
model = add_dummy_weight(model)
return model