def validate_keras_model(serialized_model, weights, dataset_iterator,
data_count):
logging.info('Keras validation just started.')
assert weights != None, "weights must not be 'None'."
model = model_from_serialized(serialized_model)
model.set_weights(weights)
history = model.evaluate_generator(dataset_iterator, steps=data_count)
metrics = dict(zip(model.metrics_names, history))
logging.info('Keras validation complete.')
return {'val_metric': metrics}
def train_keras_model(serialized_model, weights, dataset_iterator, data_count,
hyperparams, config):
logging.info('Keras training just started.')
assert weights != None, "Initial weights must not be 'None'."
model = model_from_serialized(serialized_model)
model.set_weights(weights)
hist = model.fit_generator(dataset_iterator, epochs=hyperparams['epochs'], \
steps_per_epoch=data_count//hyperparams['batch_size'])
# weights_filepath = os.path.join(
# os.path.dirname(os.path.realpath(__file__)),
# config["weights_directory"],
# uuid.uuid4().hex[:8] + ".h5"
# )
# ensure_dir(weights_filepath)
# model.save_weights(weights_filepath)
weights = model.get_weights()
logging.info('Keras training complete.')
return weights, {'training_history': hist.history}
def _initialize(self, job):
"""
Initializes and returns a DMLResult with the model
weights as specified in the model.
"""
assert job.framework_type in ['keras'], \
"Model type '{0}' is not supported.".format(job.framework_type)
logging.info("Initializing model...")
if job.framework_type == 'keras':
model = model_from_serialized(job.serialized_model)
#model.summary()
initial_weights = model.get_weights()
results = DMLResult(
status='successful',
job=job,
results={
'weights': initial_weights,
},
error_message="",
)
return results