enhancementhelp wanted
説明
Feature
Automatic mixed precision for xla has landed in pytorch 1.8.1 and torch/xla nightly.
We should enable it in create_supervised_* helper functions.
Suggested solution
Remove xla and amp checks in _check_arg().
- For
create_supervised_trainer, updatesupervised_training_step_tpu()function to acceptscalerargument just likesupervised_training_step_amp(). - For
create_supervised_evaluator, just removing xla and amp checks in_check_arg()should work. - For tests, we could remove xla checks and only run with pytorch 1.8.1.
Additional context
This feature should not be included in ignite release until the next torch and xla comes out.