def gen_model_and_train(model=None, input_shapes=None, dataset=None, visualizers=None, env_path=None, epoch_time=50,
check_point_interval_per_iter=5000):
manager = InstanceManager(env_path)
manager.gen_instance(model, input_shapes)
manager.load_visualizer(visualizers)
manager.open_tensorboard()
manager.train_model(dataset, epoch_time, check_point_interval_per_iter)
manager.close_tensorboard()
del manager
def load_model_and_train(input_shapes=None, visualizers=None, metadata_path=None, env_setting=None, dataset=None,
epoch_time=50, check_point_interval_per_iter=5000):
manager = None
try:
manager = InstanceManager(env_setting)
manager.load_model(metadata_path=metadata_path, input_shapes=input_shapes)
manager.load_visualizer(visualizers)
manager.open_tensorboard()
manager.train_model(dataset, epoch_time, check_point_interval_per_iter, is_restore=True)
except Exception as e:
print(e)
finally:
manager.close_tensorboard()
del manager