AITemplate icon indicating copy to clipboard operation
AITemplate copied to clipboard

Published 20 hours ago • verified publisher icon facebookincubator

fix bf16 lowering

Open frank-wei opened this issue 1 year ago • 1 comments

Summary:

  1. when enable bf16, torch.ops.fbgemm.generic_histogram_binning_calibration_by_feature in submod1 does not take bf16. So we need to cast its input to fp32
  2. nan_to_num could handle bf16 now

Differential Revision: D45421503

frank-wei avatar May 01 '23 16:05 frank-wei

This pull request was exported from Phabricator. Differential Revision: D45421503

facebook-github-bot avatar May 01 '23 16:05 facebook-github-bot