Port to new registration API (part 1)#36389
Port to new registration API (part 1)#36389ezyang wants to merge 16 commits intogh/ezyang/715/basefrom
Conversation
Signed-off-by: Edward Z. Yang <ezyang@fb.com> [ghstack-poisoned]
💊 Build failures summary and remediationsAs of commit 388f0c0 (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 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 74 times. |
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
Signed-off-by: Edward Z. Yang <ezyang@fb.com> ghstack-source-id: a99e00b Pull Request resolved: pytorch#36389
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
…Convolution" Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
This is a roll up of a bunch of small PRs for ease of landing. - Update reference to RegisterOperators in error message in Convolution. #36389 - Port Resize to use new registration API. #36390 - Port detach/detach to new registration API. #36512 - Add explicit schema for quantized conv/conv_prepack (fixes #36511). #36513 - Add a centralized TORCH_LIBRARY declaration for quantized and xnnpack ops (fixes #36510). #36520 - Functionalize qadd and register with new registration API. #36527 - Update quantized README for registering operators with new API. #36531 - Convert qbatch_norm to new operator registration API. #36534 - Convert qclamp to new operator registration API. #36535 - Functionalize qconcat and register with new registration API. #36536 Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
This is a roll up of a bunch of small PRs for ease of landing. - Update reference to RegisterOperators in error message in Convolution. #36389 - Port Resize to use new registration API. #36390 - Port detach/detach to new registration API. #36512 - Add explicit schema for quantized conv/conv_prepack (fixes #36511). #36513 - Add a centralized TORCH_LIBRARY declaration for quantized and xnnpack ops (fixes #36510). #36520 - Functionalize qadd and register with new registration API. #36527 - Update quantized README for registering operators with new API. #36531 - Convert qbatch_norm to new operator registration API. #36534 - Convert qclamp to new operator registration API. #36535 - Functionalize qconcat and register with new registration API. #36536 Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
This is a roll up of a bunch of small PRs for ease of landing. - Update reference to RegisterOperators in error message in Convolution. #36389 - Port Resize to use new registration API. #36390 - Port detach/detach to new registration API. #36512 - Add explicit schema for quantized conv/conv_prepack (fixes #36511). #36513 - Add a centralized TORCH_LIBRARY declaration for quantized and xnnpack ops (fixes #36510). #36520 - Functionalize qadd and register with new registration API. #36527 - Update quantized README for registering operators with new API. #36531 - Convert qbatch_norm to new operator registration API. #36534 - Convert qclamp to new operator registration API. #36535 - Functionalize qconcat and register with new registration API. #36536 ghstack-source-id: fe27bc1 Pull Request resolved: #36389
This is a roll up of a bunch of small PRs for ease of landing. - Update reference to RegisterOperators in error message in Convolution. #36389 - Port Resize to use new registration API. #36390 - Port detach/detach to new registration API. #36512 - Add explicit schema for quantized conv/conv_prepack (fixes #36511). #36513 - Add a centralized TORCH_LIBRARY declaration for quantized and xnnpack ops (fixes #36510). #36520 - Functionalize qadd and register with new registration API. #36527 - Update quantized README for registering operators with new API. #36531 - Convert qbatch_norm to new operator registration API. #36534 - Convert qclamp to new operator registration API. #36535 - Functionalize qconcat and register with new registration API. #36536 Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
This is a roll up of a bunch of small PRs for ease of landing. - Update reference to RegisterOperators in error message in Convolution. #36389 - Port Resize to use new registration API. #36390 - Port detach/detach to new registration API. #36512 - Add explicit schema for quantized conv/conv_prepack (fixes #36511). #36513 - Add a centralized TORCH_LIBRARY declaration for quantized and xnnpack ops (fixes #36510). #36520 - Functionalize qadd and register with new registration API. #36527 - Update quantized README for registering operators with new API. #36531 - Convert qbatch_norm to new operator registration API. #36534 - Convert qclamp to new operator registration API. #36535 - Functionalize qconcat and register with new registration API. #36536 Differential Revision: [D20964193](https://our.internmc.facebook.com/intern/diff/D20964193) [ghstack-poisoned]
| ('quantized::batch_norm', datetime.date(2020, 4, 20)), | ||
| ('aten::sizes', datetime.date(2020, 4, 30)), | ||
| ('aten::strides', datetime.date(2020, 4, 30)), | ||
| ('quantized::conv_prepack', datetime.date(2020, 6, 1)), |
There was a problem hiding this comment.
what's going on here? why do you need to put it into BC test?
There was a problem hiding this comment.
Oh, this is mistake from when I typoed conv_prepack for an registration, and that lit up the bc check, and then when I fixed the typo I didn't fix the registration
There was a problem hiding this comment.
I need this in the BC test because I made a (technical) BC breaking change.
Previously, conv_prepack was registered with inferred schema:
- .op("quantized::conv_prepack", // conv_prepack is deprecated, please use
- // conv2d_prepack for 2D conv.
With this PR, it is now done with explicit schema:
+ // conv_prepack is deprecated, please use conv2d_prepack for 2D conv.
+ m.def("conv_prepack(Tensor weight, Tensor? bias, int[] stride, int[] padding, int[] dilation, int groups) -> Tensor");
Moving from implicit to explicit is considered bc breaking by the bc script, thus I must whitelist it.
|
Appears to have caused the following in Reverting. |
|
Stunning, the local ci was all green! XD |
Signed-off-by: Edward Z. Yang <ezyang@fb.com> ghstack-source-id: a99e00b Pull Request resolved: pytorch/pytorch#36389
…pytorch#36389) Summary: Pull Request resolved: pytorch#36389 Signed-off-by: Edward Z. Yang <ezyang@fb.com> Test Plan: Imported from OSS Differential Revision: D20964193 Pulled By: ezyang fbshipit-source-id: 27aeea01ccf5dfcebb8f043cde009a14dde3958e
Stack from ghstack:
This is a roll up of a bunch of small PRs for ease of landing.
Differential Revision: D20964193