Skip to content
This repository was archived by the owner on Jul 7, 2023. It is now read-only.
This repository was archived by the owner on Jul 7, 2023. It is now read-only.

Python 3 compatibility issue with self_generated_targets #1298

@ywkim

Description

@ywkim

Description

I get AssertionError when trying to return self_generated_targets in a model.

Environment information

OS: macOS 10.13.4

$ pip freeze | grep tensor
mesh-tensorflow==0.0.4
tensor2tensor==1.11.0
tensorboard==1.12.0
tensorflow==1.12.0
tensorflow-metadata==0.9.0
tensorflow-probability==0.5.0

$ python -V
Python 3.6.4

For bugs: error logs

  File "/Users/ywkim/tensor2tensor/tensor2tensor/utils/t2t_model.py", line 1266, in wrapping_model_fn                                                                 
    use_tpu=use_tpu)
  File "/Users/ywkim/tensor2tensor/tensor2tensor/utils/t2t_model.py", line 1340, in estimator_model_fn                                                                
    "Found {}".format(logits.keys())
AssertionError: Expect only key 'logits' when there is 'self_generated_targets'. Found dict_keys(['logits'])

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions