squeeze: allow squeezing multiple dimensions at once #89017

peterbell10 · 2022-11-14T21:49:08Z

Stack from ghstack (oldest at bottom):

This addresses part 1 of the issue, allowing torch.squeeze to be
passed a tuple of dimensions. e.g.

x.squeeze(0).squeeze(0)

can now be written

x.squeeze((0, 1))

(assuming x has at least 2 dimensions)

cc @mlazos @soumith @voznesenskym @yanboliang @penguinwu @anijain2305 @EikanWang @jgong5 @Guobing-Chen @chunyuan-w @XiaobingSuper @zhuhaozhe @blzheng @Xia-Weiwen @wenzhe-nrv @jiayisunx @desertfire

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze(0, 1) ``` (assuming x has at least 2 dimensions) [ghstack-poisoned]

pytorch-bot · 2022-11-14T21:49:12Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/89017

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit e44436d:
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze(0, 1) ``` (assuming x has at least 2 dimensions) ghstack-source-id: 23ef534 Pull Request resolved: #89017

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze(0, 1) ``` (assuming x has at least 2 dimensions) [ghstack-poisoned]

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze(0, 1) ``` (assuming x has at least 2 dimensions) ghstack-source-id: 8b633ca Pull Request resolved: #89017

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze(0, 1) ``` (assuming x has at least 2 dimensions) [ghstack-poisoned]

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze(0, 1) ``` (assuming x has at least 2 dimensions) ghstack-source-id: e65942c Pull Request resolved: #89017

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) [ghstack-poisoned]

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) ghstack-source-id: c486073 Pull Request resolved: #89017

albanD · 2022-11-22T21:54:44Z

@mruberry can you take care of this one?

…dimensions at once" Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) [ghstack-poisoned]

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) ghstack-source-id: 9e7299c Pull Request resolved: #89017

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) [ghstack-poisoned]

…ns at once" Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) cc mlazos soumith voznesenskym yanboliang penguinwu anijain2305 EikanWang jgong5 Guobing-Chen chunyuan-w XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx desertfire [ghstack-poisoned]

albanD

Ok!

peterbell10 · 2023-01-04T14:35:55Z

@pytorchbot merge -f "ROCm timeout is unrelated"

pytorchmergebot · 2023-01-04T14:40:51Z

Merge started

Your change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

mehtanirav · 2023-01-05T19:20:07Z

This PR is causing internal build failures with errors like

[2023-01-04T23:03:00.073-08:00] Stderr: fbcode/caffe2/aten/src/ATen/native/nested/NestedTensorMath.cpp:711:20: error: comparison of integers of different signs: 'int64_t' (aka 'long') and 'unsigned long' [-Werror,-Wsign-compare]
[CONTEXT] [2023-01-04T23:03:00.073-08:00]   TORCH_CHECK(ndim > 1 + dims.size(),
[CONTEXT] [2023-01-04T23:03:00.073-08:00]               ~~~~ ^ ~~~~~~~~~~~~~~~

mehtanirav · 2023-01-05T19:20:48Z

@pytorchbot revert -m "Internal breakages" -c ghfirst

pytorchmergebot · 2023-01-05T19:25:01Z

@pytorchbot successfully started a revert job. Check the current status here.
Questions? Feedback? Please reach out to the PyTorch DevX Team

pytorchmergebot · 2023-01-05T19:25:13Z

@peterbell10 your PR has been successfully reverted.

This reverts commit e26cb06. Reverted #89017 on behalf of https://github.com/mehtanirav due to Internal breakages

ngimel · 2023-01-05T19:27:05Z

Next PR in the stack will also need to be reverted, why wasn't it reverted automatically?

mehtanirav · 2023-01-05T19:33:29Z

Next PR in the stack will also need to be reverted, why wasn't it reverted automatically?

Looks like there was no conflict in reverting this PR independently. do we still need to revert #91602?

ngimel · 2023-01-05T19:48:53Z

Yes, even though there's no conflicts, #91602 relies on functionality introduced here. Could we have done a quick fix here instead? If unsigned comparison is the only problem, that's an easy forward fix.

Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) cc mlazos soumith voznesenskym yanboliang penguinwu anijain2305 EikanWang jgong5 Guobing-Chen chunyuan-w XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx desertfire [ghstack-poisoned]

…t once" Ref #70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) cc mlazos soumith voznesenskym yanboliang penguinwu anijain2305 EikanWang jgong5 Guobing-Chen chunyuan-w XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx desertfire [ghstack-poisoned]

Ref pytorch#70924 This addresses part 1 of the issue, allowing `torch.squeeze` to be passed a tuple of dimensions. e.g. ```python x.squeeze(0).squeeze(0) ``` can now be written ```python x.squeeze((0, 1)) ``` (assuming x has at least 2 dimensions) ghstack-source-id: c8cb529 Pull Request resolved: pytorch#89017

peterbell10 · 2023-01-17T14:18:32Z

@pytorchbot merge

pytorchmergebot · 2023-01-17T14:20:11Z

Merge started

Your 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

Advanced Debugging

Check the merge workflow status
here

pytorchbot added the open source label Nov 14, 2022

peterbell10 added release notes: python_frontend python frontend release notes category topic: improvements topic category labels Nov 15, 2022

peterbell10 marked this pull request as ready for review November 20, 2022 21:30

peterbell10 requested review from albanD, mruberry, ngimel, soulitzer and zou3519 as code owners November 20, 2022 21:30

albanD approved these changes Jan 3, 2023

View reviewed changes

peterbell10 added the ciflow/trunk Trigger trunk jobs on your pull request label Jan 3, 2023

pytorchmergebot added the Merged label Jan 4, 2023

pytorchmergebot closed this in e26cb06 Jan 4, 2023

pytorchmergebot added a commit that referenced this pull request Jan 5, 2023

Revert "squeeze: allow squeezing multiple dimensions at once (#89017)"

df4b3b1

This reverts commit e26cb06. Reverted #89017 on behalf of https://github.com/mehtanirav due to Internal breakages

pytorchmergebot added the Reverted label Jan 5, 2023

peterbell10 reopened this Jan 6, 2023

peterbell10 added 3 commits January 6, 2023 19:25

pytorchmergebot closed this in fb1427e Jan 17, 2023

facebook-github-bot deleted the gh/peterbell10/464/head branch June 8, 2023 18:22

ScXfjiang mentioned this pull request Apr 22, 2024

Dev quant squeeze Xilinx/brevitas#941

Closed

2 tasks

github-actions bot mentioned this pull request Apr 25, 2024

py3-scipy/1.13.0 package update wolfi-dev/os#16208

Merged

squeeze: allow squeezing multiple dimensions at once #89017

squeeze: allow squeezing multiple dimensions at once #89017

Uh oh!

Conversation

peterbell10 commented Nov 14, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Nov 14, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/89017

✅ No Failures

Uh oh!

albanD commented Nov 22, 2022

Uh oh!

albanD left a comment

Choose a reason for hiding this comment

Uh oh!

peterbell10 commented Jan 4, 2023

Uh oh!

pytorchmergebot commented Jan 4, 2023

Merge started

Uh oh!

mehtanirav commented Jan 5, 2023

Uh oh!

mehtanirav commented Jan 5, 2023

Uh oh!

pytorchmergebot commented Jan 5, 2023

Uh oh!

pytorchmergebot commented Jan 5, 2023

Uh oh!

ngimel commented Jan 5, 2023

Uh oh!

mehtanirav commented Jan 5, 2023

Uh oh!

ngimel commented Jan 5, 2023

Uh oh!

peterbell10 commented Jan 17, 2023

Uh oh!

pytorchmergebot commented Jan 17, 2023

Merge started

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

peterbell10 commented Nov 14, 2022 •

edited

Loading

pytorch-bot bot commented Nov 14, 2022 •

edited

Loading