Modify signature of dequantize ops for decomposed quantized Tensor #2308
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/2308
Note: Links to docs will display an error until the docs builds have been completed. ❌ 6 New FailuresAs of commit 91acd1d with merge base 1fb447e ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
facebook-github-bot
added
CLA Signed
manuelcandales
pushed a commit
to manuelcandales/pytorch
that referenced
this pull request
Mar 7, 2024
…ytorch#121450) Summary: Pull Request resolved: pytorch#121450 X-link: pytorch/executorch#2308 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel Pull Request resolved: pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
d28fe4a to
6d4080e
Compare
manuelcandales
pushed a commit
to manuelcandales/executorch-1
that referenced
this pull request
Mar 8, 2024
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
manuelcandales
pushed a commit
to manuelcandales/executorch-1
that referenced
this pull request
Mar 8, 2024
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
6d4080e to
41c27a2
Compare
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
manuelcandales
pushed a commit
to manuelcandales/executorch-1
that referenced
this pull request
Mar 8, 2024
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
41c27a2 to
fbe1c8c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
manuelcandales
pushed a commit
to manuelcandales/executorch-1
that referenced
this pull request
Mar 8, 2024
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
fbe1c8c to
3068fbb
Compare
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
jerryzh168
approved these changes
Mar 8, 2024
manuelcandales
pushed a commit
to manuelcandales/executorch-1
that referenced
this pull request
Mar 8, 2024
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
3068fbb to
b74bcf3
Compare
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
manuelcandales
pushed a commit
to manuelcandales/executorch-1
that referenced
this pull request
Mar 8, 2024
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
b74bcf3 to
2dac562
Compare
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
manuelcandales
pushed a commit
to manuelcandales/pytorch
that referenced
this pull request
Mar 11, 2024
…ytorch#121450) Summary: X-link: pytorch/executorch#2308 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
2dac562 to
72cc9e0
Compare
manuelcandales
pushed a commit
to manuelcandales/executorch-1
that referenced
this pull request
Mar 11, 2024
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
72cc9e0 to
722d09c
Compare
manuelcandales
pushed a commit
to manuelcandales/executorch-1
that referenced
this pull request
Mar 12, 2024
…ytorch#2308) Summary: X-link: pytorch/pytorch#121450 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel X-link: pytorch/pytorch#119173 Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
manuelcandales
force-pushed
the
export-D53590486
branch
from
722d09c to
91acd1d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
|
This pull request was exported from Phabricator. Differential Revision: D53590486 |
manuelcandales
pushed a commit
to manuelcandales/pytorch
that referenced
this pull request
Mar 12, 2024
…ytorch#121450) Summary: X-link: pytorch/executorch#2308 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales
|
@manuelcandales merged this pull request in 969638c. |
pytorchmergebot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Mar 12, 2024
…119173) (#121450) Summary: X-link: pytorch/executorch#2308 Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any. At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization. This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead. cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel Reviewed By: digantdesai Differential Revision: D53590486 Pulled By: manuelcandales Co-authored-by: kausik <[email protected]> Pull Request resolved: #121450 Approved by: https://github.com/jerryzh168
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.
Summary:
Note: The initial purpose of this PR is to draw suggestion and feedback regarding better alternative, if any.
At present, dequantize op for decomposed quantized Tensor representation e.g. dequantize_per_tensor() assumes the output dtype as torch.float and hence, it does not have the output dtype in its operator argument list. However, this op signature becomes unusable when the assumption breaks. Because, in case the output dtype is different from torch.float, there is no way to specify the same during dequantization.
This change is aimed at generalizing the signature of dequantize op like dequantize_per_tensor() for wider use-cases where the output dtype can be different from torch.float and needs to passed during dequantization. The proposal is to use an additional argument named 'output_dtype' to solve the problem. However, we would also like to have suggestion and feedback regarding any better alternative that can be used instead.
cc jerryzh168 jianyuh raghuramank100 jamesr66a vkuzo jgong5 Xia-Weiwen leslie-fang-intel
X-link: pytorch/pytorch#119173
Reviewed By: digantdesai
Differential Revision: D53590486
Pulled By: manuelcandales