Auto Lazy-Reorder for BF16 #30
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Conflicts: torch_ipex/csrc/aten_ipex_bridge.cpp
…use it out of scope
TODO: Move BF16 reorder to DNNL OPs
EikanWang
pushed a commit
that referenced
this pull request
Oct 4, 2021
* linear weight prepack cache for jit path * add description of weight prepack * fix conv bn fusion error when some model can't use FX * change the description of weight prepack
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
IPEX will enable Lazy-Reorder for BF16. The user just needs to call
enable_mix_bf16_fp32to enable BF16 mix precision. The user does not need to callxxx.to(torch.bfloat16)/xxx.bfloat16()explicitly.IPEX will auto-reorder the input tensor to BF16 if the OP supports BF16, otherwise, it will be reordered to float.