[quant] Implement APoT_tensor class #79940
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
[ghstack-poisoned]
This was referenced Jun 21, 2022
🔗 Helpful links
✅ No Failures (0 Pending)As of commit ed16153 (more details on the Dr. CI page): Expand to see more💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Please report bugs/suggestions to the (internal) Dr. CI Users group. |
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
HDCharles
reviewed
Jun 22, 2022
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
dzdang
approved these changes
Jun 22, 2022
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and has a method to convert APoT tensor to int representation. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
asl3
added
topic: not user facing
### Summary: This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and a method to return the int representation of an APoT tensor. ### Test Plan: Run unit tests with: `python pytorch/test/quantization/core/experimental/test_quantized_tensor.py` [ghstack-poisoned]
|
@pytorchbot merge -g |
|
@pytorchbot successfully started a merge job. Check the current status here |
dzdang
reviewed
Jun 22, 2022
|
|
||
| # class to store APoT quantized tensor | ||
| class TensorAPoT(torch.Tensor): | ||
| class TensorAPoT(): |
There was a problem hiding this comment.
why is TensorAPoT no longer inheriting from torch.Tensor?
There was a problem hiding this comment.
When TensorAPoT inherits from torch.Tensor I get an issue with my TensorAPoT init function since I am not passing in the expected args:
`TypeError: new() received an invalid combination of arguments - got (Tensor, int, int, bool), but expected one of:
- (*, torch.device device)
- (torch.Storage storage)
- (Tensor other)
- (tuple of ints size, *, torch.device device)
- (object data, *, torch.device device)`
I didn't think any of these made sense for my TensorAPoT object since I am passing in a quantizer object. Closest was the (Tensor other) option, but in that case I would need to create the quantizer in my init function and would need to pass in the args needed by quantizer init , which would also violate the expected args.
Is it okay if TensorAPoT doesn't inherit from torch.Tensor?
There was a problem hiding this comment.
I feel this is fine for now, this might involve a bit more design about how do we allow people to extend quantized tensor in python, that's something we can explore in H2
|
@asl3 your PR has been successfully merged. |
aovladi
pushed a commit
that referenced
this pull request
Jun 22, 2022
Pull Request resolved: #79940 Approved by: https://github.com/dzdang
facebook-github-bot
pushed a commit
that referenced
this pull request
Jun 23, 2022
Summary: Pull Request resolved: #79940 Approved by: https://github.com/dzdang Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/82a1961129bc682da2da9863d88d39525f4301b6 Reviewed By: atalman Differential Revision: D37357398 Pulled By: asl3 fbshipit-source-id: 89d962cddd2f2b2feb96ace3d3b618a7a23a4fa0
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
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.
Stack from ghstack (oldest at bottom):
Summary:
This PR implements functionality to store APoT tensors. The APoT tensor class contains a quantizer and a method to return the int representation of an APoT tensor.
Test Plan:
Run unit tests with:
python pytorch/test/quantization/core/experimental/test_quantized_tensor.py