Make Float8 types serializeable #114662
Closed
Make Float8 types serializeable #114662
Conversation
This file contains 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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/114662
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 79e4b18 with merge base b6a30bb ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
ngimel
reviewed
Nov 28, 2023
torch/_tensor.py
Outdated
| # need to wrap with TypedStorage | ||
| args = ( | ||
| torch.storage.TypedStorage( | ||
| v3_dtypes = [torch.float8_e5m2, torch.float8_e4m3fn] |
There was a problem hiding this comment.
Probably makes sense to serialize all tensors whose dtype is not in the dict like this. No bc breaking, as those tensors couldn't have been serialized at all.
There was a problem hiding this comment.
Hmm, so do the inverse. Makes sense, will do in follow up PR (and probably past the branch cut, as it feels a bit risky)
malfet
added
enhancement
By finally breaking FC promise on new dtypes by serializing untyped
storage and tensor dtypes
- Add `_rebuild_tensor_v3` that takes an extra dtype argument
- In `Tensor.__reduce_ex__` serialize tensor using untyped storage for
v3_dtypes (which are at the moment limited to float8 dtypes)
Test plan: `python -c "import torch;x=torch.arange(10).to(dtype=torch.float8_e4m3fn);torch.save(x, 'pt.pt');print(torch.load('pt.pt'))"`
malfet
force-pushed
the
malfet/make-f8-serializable
branch
from
c48da22
to
79e4b18
Compare
ngimel
approved these changes
Nov 29, 2023
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
hyperfraise
pushed a commit
to hyperfraise/pytorch
that referenced
this pull request
Dec 21, 2023
By finally breaking FC promise on new dtypes by serializing untyped
storage and tensor dtypes
- Add `_rebuild_tensor_v3` that takes an extra dtype argument
- In `Tensor.__reduce_ex__` serialize tensor using untyped storage for
v3_dtypes (which are at the moment limited to float8 dtypes)
Test plan: `python -c "import torch;x=torch.arange(10).to(dtype=torch.float8_e4m3fn);torch.save(x, 'pt.pt');print(torch.load('pt.pt'))"`
Fixes pytorch#114634
Pull Request resolved: pytorch#114662
Approved by: https://github.com/ngimel
hyperfraise
pushed a commit
to hyperfraise/pytorch
that referenced
this pull request
Dec 21, 2023
By finally breaking FC promise on new dtypes by serializing untyped
storage and tensor dtypes
- Add `_rebuild_tensor_v3` that takes an extra dtype argument
- In `Tensor.__reduce_ex__` serialize tensor using untyped storage for
v3_dtypes (which are at the moment limited to float8 dtypes)
Test plan: `python -c "import torch;x=torch.arange(10).to(dtype=torch.float8_e4m3fn);torch.save(x, 'pt.pt');print(torch.load('pt.pt'))"`
Fixes pytorch#114634
Pull Request resolved: pytorch#114662
Approved by: https://github.com/ngimel
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.
By finally breaking FC promise on new dtypes by serializing untyped
storage and tensor dtypes
_rebuild_tensor_v3that takes an extra dtype argumentTensor.__reduce_ex__serialize tensor using untyped storage forv3_dtypes (which are at the moment limited to float8 dtypes)
Test plan:
python -c "import torch;x=torch.arange(10).to(dtype=torch.float8_e4m3fn);torch.save(x, 'pt.pt');print(torch.load('pt.pt'))"Fixes #114634