tvalentyn commented on code in PR #21819:
URL: https://github.com/apache/beam/pull/21819#discussion_r895734980
##########
sdks/python/apache_beam/examples/inference/pytorch_image_classification.py:
##########
@@ -114,10 +115,13 @@ def run(argv=None, model_class=None, model_params=None,
save_main_session=True):
model_class = MobileNetV2
model_params = {'num_classes': 1000}
- model_loader = PytorchModelLoader(
- state_dict_path=known_args.model_state_dict_path,
- model_class=model_class,
- model_params=model_params)
+ # the input to RunInference transform is keyed. Wrap
+ # PytorchModelHandler on KeyedModelHandler for keyed examples.
+ model_loader = KeyedModelHandler(
+ PytorchModelHandler(
+ state_dict_path=known_args.model_state_dict_path,
+ model_class=model_class,
+ model_params=model_params))
Review Comment:
ok. I thought the splitting of model handlers would allow us getting the
type deterministically instead of having to make a union. we can look into it
separately.
cc: @robertwb @yeandy @rezarokni
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@beam.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
