Add guard elimination cases for operators encountered on an RL workload.#34967
Add guard elimination cases for operators encountered on an RL workload.#34967resistor wants to merge 1 commit intopytorch:masterfrom
Conversation
💊 CircleCI build failures summary and remediationsAs of commit a31cb90 (more details on the Dr. CI page): ✅ None of the build failures appear to be your fault 💚
🚧 2 upstream failures:These were probably caused by upstream breakages:
This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions on the GitHub issue tracker. This comment has been revised 4 times. |
There was a problem hiding this comment.
do we know what stacklevel does?
There was a problem hiding this comment.
The output shape is always the same as the input shape: https://pytorch.org/docs/stable/nn.html#torch.nn.Softmax2d
There was a problem hiding this comment.
the 6th argument of convXd seems to be the number of groups? Shouldn't that affect output shapes even if input and weights' shapes are fixed?
There was a problem hiding this comment.
Groups affects the result values but not the shape.
facebook-github-bot
left a comment
There was a problem hiding this comment.
@resistor has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
…ad. (pytorch#34967) Summary: Pull Request resolved: pytorch#34967 Differential Revision: D20547478 Pulled By: resistor fbshipit-source-id: da7df159fd6098d0f1278b8088bbbe6717b79cfc
No description provided.