{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":65600975,"defaultBranch":"main","name":"pytorch","ownerLogin":"pytorch","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2016-08-13T05:26:41.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/21003710?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1718205294.0","currentOid":""},"activityList":{"items":[{"before":"22379fa6e9ea53e7e2c943aaa6f084327d537dfd","after":"187d216de634676169408f14be4b0bf9ff86bdfd","ref":"refs/heads/gh/amjames/21/head","pushedAt":"2024-06-12T15:15:40.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"amjames","name":"Andrew James","path":"/amjames","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9041474?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":"8f8f2f9dcee909a78189bff8a760986da2ea1c5e","after":"eb2b87e6d9d851a7aa00d9c0e266bf676fed6247","ref":"refs/heads/gh/amjames/24/head","pushedAt":"2024-06-12T15:15:40.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"amjames","name":"Andrew James","path":"/amjames","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9041474?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":"a5365398afeae100891e1df70937e16c79982ce6","after":"ae548318a953f64c8f7e2390f48246a18f4952f7","ref":"refs/heads/gh/amjames/24/base","pushedAt":"2024-06-12T15:15:38.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"amjames","name":"Andrew James","path":"/amjames","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9041474?s=80&v=4"},"commit":{"message":"Update (base update)\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update (base update)"}},{"before":"606d0137c8e0987b590785067cd84f24cf75402f","after":null,"ref":"refs/tags/ciflow/trunk/125795","pushedAt":"2024-06-12T15:14:53.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}},{"before":"441cb2441cdb4250a7d73d8dc7a73e7aba6e382e","after":"053d42e83c7e46293abdfb84a505cab0fc0d029e","ref":"refs/heads/gh/soulitzer/304/orig","pushedAt":"2024-06-12T15:14:50.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"soulitzer","name":"Jeffrey Wan","path":"/soulitzer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/13428986?s=80&v=4"},"commit":{"message":"[checkpoint] Clean up selective activation checkpoint and make public\n\nghstack-source-id: 95f2b21b5ae56e84108639664662388bdb9ea243\nPull Request resolved: https://github.com/pytorch/pytorch/pull/125795","shortMessageHtmlLink":"[checkpoint] Clean up selective activation checkpoint and make public"}},{"before":"c5427a8ae5647eb3484938d2f87c0d5723584e02","after":"d3820a85dcb398a10565c9a0e1dfbb7b988af36b","ref":"refs/heads/gh/soulitzer/306/orig","pushedAt":"2024-06-12T15:14:50.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"soulitzer","name":"Jeffrey Wan","path":"/soulitzer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/13428986?s=80&v=4"},"commit":{"message":"Support nested tensor with activation checkpoint\n\nghstack-source-id: a2cedfb9a75526b9cda678006c657805eede972b\nPull Request resolved: https://github.com/pytorch/pytorch/pull/127959","shortMessageHtmlLink":"Support nested tensor with activation checkpoint"}},{"before":"046ca667046bcd1ae077877319a32862c51cda6a","after":"284195cb26523b456aed74f614fd2dc83391296b","ref":"refs/heads/gh/soulitzer/306/head","pushedAt":"2024-06-12T15:14:47.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"soulitzer","name":"Jeffrey Wan","path":"/soulitzer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/13428986?s=80&v=4"},"commit":{"message":"Update on \"[checkpoint] Activation checkpoint preserves object identity when tensor does not require grad\"\n\n\r\nThis PR ensures that if a tensor does not require grad, and it is saved for backward, (1) unpacking in general preserve the object, (2) non-reentrant checkpoint preserves the object.\r\n\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update on \"[checkpoint] Activation checkpoint preserves object identi…"}},{"before":"606d0137c8e0987b590785067cd84f24cf75402f","after":"860eab309d42c5c958b423325395b2fa9343af74","ref":"refs/heads/gh/soulitzer/304/head","pushedAt":"2024-06-12T15:14:47.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"soulitzer","name":"Jeffrey Wan","path":"/soulitzer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/13428986?s=80&v=4"},"commit":{"message":"Update on \"[checkpoint] Clean up selective activation checkpoint and make public\"\n\n\r\nRelated doc: https://docs.google.com/document/d/1BKyizkZPdri9mHqdDOLAUpkI7SbbKfLHRFVVpK9ZWqo/edit\r\n\r\nMemory considerations:\r\n- As with the existing SAC, cached values are cleared upon first use. \r\n- We error if the user wishes to backward a second time on a region forwarded with SAC enabled.\r\n\r\nIn-place:\r\n- We use version counting to enforce that if any cached tensor has been mutated. In-place operations not mutating cached tensors are allowed.\r\n- `allow_cache_entry_mutation=True` can be passed to disable this check (useful in the case of auto AC where the user is cleverly also saves the output of the in-place)\r\n\r\nRandomness, views\r\n- Currently in this PR, we don't do anything special for randomness or views, the author of the policy function is expected to handle them properly. (Would it would be beneficial to error? - we either want to save all or recompute all random tensors)\r\n\r\nTensor object preservation\r\n- We guarantee that if a tensor does not requires grad, and it is saved, then what you get out is the same tensor object. If the tensor does require grad, we must detach to avoid creating a reference cycle. This is a nice guarantee for nested tensors which care about the object identity of of the offsets tensor.\r\n\r\nPolicy function \r\n- Enum values are `{MUST,PREFER}_{SAVE,RECOMPUTE}` (bikeshed welcome). Alternatively there was `{SAVE,RECOMPUTE}_{NON_,}OVERRIDABLE`. The former was preferred bc it seemed clearer that two `MUST` clashing should error, versus it is ambiguous whether two `NON_OVERRIDABLE` being stacked should silently ignore or error.\r\n- The usage of Enum today. There actually is NO API to stack SAC policies today. The only thing the Enum should matter for in the near term is the compiler. The stacking SAC policy would be useful if someone wants to implement something like simple FSDP, but it is not perfect because with a policy of `PREFER_SAVE` you are actually saving more than autograd would save normally (would be fixed with AC v3).\r\n- The number of times we call the policy_fn is something documented part of public API. We call the policy function for all ops except detach because detach is itself called a different number of times by AC between forward and recompute. \r\n- The policy function can be a stateful object (we do NOT make separate copies of this object for forward/recompute, the user is expected to handle that via is_recompute see below).\r\nTensors guaranteed to be the same tensor as-is\r\n- Policy function signature takes ctx object as its first argument. The ctx function is an object encapsulating info that may be useful to the user, it currently only holds \"is_recompute\". Adding this indirection gives us flexibility to add more attrs later if necessary.\r\n\r\n\"bc-breaking\" for existing users of the private API:\r\n- Existing policy functions must now change their return value to use the Enum.\r\n- Existing calls to `_pt2_selective_checkpoint_context_fn_gen` must be renamed to `gen_selective_checkpoint_context_fn`. The way you use the API remains the same. It would've been nice to do something different (not make the user have to use functools.partial?), but this was the easiest to compile (idk if this should actually be a constraint). \r\n\r\ncc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update on \"[checkpoint] Clean up selective activation checkpoint and …"}},{"before":"83bdeaffd0afe248ec8bdfb37587e99a65d3aec7","after":"03f1b0ee93bbc01157f1451406184335bedf7fc2","ref":"refs/heads/gh/soulitzer/306/base","pushedAt":"2024-06-12T15:14:44.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"soulitzer","name":"Jeffrey Wan","path":"/soulitzer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/13428986?s=80&v=4"},"commit":{"message":"Update base for Update on \"[checkpoint] Activation checkpoint preserves object identity when tensor does not require grad\"\n\n\r\nThis PR ensures that if a tensor does not require grad, and it is saved for backward, (1) unpacking in general preserve the object, (2) non-reentrant checkpoint preserves the object.\r\n\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update base for Update on \"[checkpoint] Activation checkpoint preserv…"}},{"before":"8eab822cb33e887d01727cfd2e0a51f78b4a7da4","after":"d2cb0bed29e7d25cdbf262180a31459919e21c54","ref":"refs/heads/gh/soulitzer/304/base","pushedAt":"2024-06-12T15:14:44.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"soulitzer","name":"Jeffrey Wan","path":"/soulitzer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/13428986?s=80&v=4"},"commit":{"message":"Update base for Update on \"[checkpoint] Clean up selective activation checkpoint and make public\"\n\n\r\nRelated doc: https://docs.google.com/document/d/1BKyizkZPdri9mHqdDOLAUpkI7SbbKfLHRFVVpK9ZWqo/edit\r\n\r\nMemory considerations:\r\n- As with the existing SAC, cached values are cleared upon first use. \r\n- We error if the user wishes to backward a second time on a region forwarded with SAC enabled.\r\n\r\nIn-place:\r\n- We use version counting to enforce that if any cached tensor has been mutated. In-place operations not mutating cached tensors are allowed.\r\n- `allow_cache_entry_mutation=True` can be passed to disable this check (useful in the case of auto AC where the user is cleverly also saves the output of the in-place)\r\n\r\nRandomness, views\r\n- Currently in this PR, we don't do anything special for randomness or views, the author of the policy function is expected to handle them properly. (Would it would be beneficial to error? - we either want to save all or recompute all random tensors)\r\n\r\nTensor object preservation\r\n- We guarantee that if a tensor does not requires grad, and it is saved, then what you get out is the same tensor object. If the tensor does require grad, we must detach to avoid creating a reference cycle. This is a nice guarantee for nested tensors which care about the object identity of of the offsets tensor.\r\n\r\nPolicy function \r\n- Enum values are `{MUST,PREFER}_{SAVE,RECOMPUTE}` (bikeshed welcome). Alternatively there was `{SAVE,RECOMPUTE}_{NON_,}OVERRIDABLE`. The former was preferred bc it seemed clearer that two `MUST` clashing should error, versus it is ambiguous whether two `NON_OVERRIDABLE` being stacked should silently ignore or error.\r\n- The usage of Enum today. There actually is NO API to stack SAC policies today. The only thing the Enum should matter for in the near term is the compiler. The stacking SAC policy would be useful if someone wants to implement something like simple FSDP, but it is not perfect because with a policy of `PREFER_SAVE` you are actually saving more than autograd would save normally (would be fixed with AC v3).\r\n- The number of times we call the policy_fn is something documented part of public API. We call the policy function for all ops except detach because detach is itself called a different number of times by AC between forward and recompute. \r\n- The policy function can be a stateful object (we do NOT make separate copies of this object for forward/recompute, the user is expected to handle that via is_recompute see below).\r\nTensors guaranteed to be the same tensor as-is\r\n- Policy function signature takes ctx object as its first argument. The ctx function is an object encapsulating info that may be useful to the user, it currently only holds \"is_recompute\". Adding this indirection gives us flexibility to add more attrs later if necessary.\r\n\r\n\"bc-breaking\" for existing users of the private API:\r\n- Existing policy functions must now change their return value to use the Enum.\r\n- Existing calls to `_pt2_selective_checkpoint_context_fn_gen` must be renamed to `gen_selective_checkpoint_context_fn`. The way you use the API remains the same. It would've been nice to do something different (not make the user have to use functools.partial?), but this was the easiest to compile (idk if this should actually be a constraint). \r\n\r\ncc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update base for Update on \"[checkpoint] Clean up selective activation…"}},{"before":"f3345692942d6c57f06c149d3459892c12b35e24","after":null,"ref":"refs/tags/ciflow/mps/127231","pushedAt":"2024-06-12T15:13:13.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}},{"before":"f3345692942d6c57f06c149d3459892c12b35e24","after":null,"ref":"refs/tags/ciflow/inductor/127231","pushedAt":"2024-06-12T15:13:12.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}},{"before":"d4f1954fb419c28799d916881ea3b3c2d22e69dd","after":"6bb44443844fa10938fdaec8194be1a1680f13c1","ref":"refs/heads/gh/rec/8/orig","pushedAt":"2024-06-12T15:13:08.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"rec","name":"Tom Ritchford","path":"/rec","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/253362?s=80&v=4"},"commit":{"message":"Add OpInfo entry for as_strided_copy\n\nghstack-source-id: c0dbb53ac1e05d603b68cc9e5040663607f36e24\nPull Request resolved: https://github.com/pytorch/pytorch/pull/127231","shortMessageHtmlLink":"Add OpInfo entry for as_strided_copy"}},{"before":"f3345692942d6c57f06c149d3459892c12b35e24","after":"88b0811f6cdec48e4dff2e165b89d9621d3c9fe5","ref":"refs/heads/gh/rec/8/head","pushedAt":"2024-06-12T15:13:05.000Z","pushType":"push","commitsCount":6,"pusher":{"login":"rec","name":"Tom Ritchford","path":"/rec","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/253362?s=80&v=4"},"commit":{"message":"Update on \"Add OpInfo entry for as_strided_copy\"\n\n\n\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update on \"Add OpInfo entry for as_strided_copy\""}},{"before":"bdf8802b101ba5d67d83e49fb21a57d5613c0fee","after":"c4321b6a7d593b7602d0e704f6ea4ba081150893","ref":"refs/heads/gh/rec/8/base","pushedAt":"2024-06-12T15:13:01.000Z","pushType":"push","commitsCount":5,"pusher":{"login":"rec","name":"Tom Ritchford","path":"/rec","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/253362?s=80&v=4"},"commit":{"message":"Update base for Update on \"Add OpInfo entry for as_strided_copy\"\n\n\n\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update base for Update on \"Add OpInfo entry for as_strided_copy\""}},{"before":"a0b150a40c6e11f8d8be0b0a2ccf1ebb61499473","after":null,"ref":"refs/tags/ciflow/inductor/127836","pushedAt":"2024-06-12T15:06:51.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}},{"before":"46223e7e480132dc1881504b7687f79ec1abda1c","after":"f4a458d0798ee7d572941732530bf2795cf34591","ref":"refs/heads/gh/aorenste/56/orig","pushedAt":"2024-06-12T15:06:46.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"aorenste","name":"Aaron Orenstein","path":"/aorenste","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/19294870?s=80&v=4"},"commit":{"message":"Flip default value for mypy disallow_untyped_defs\n\nghstack-source-id: 871717926b244416132a0b731c143932772ddd96\nPull Request resolved: https://github.com/pytorch/pytorch/pull/127836","shortMessageHtmlLink":"Flip default value for mypy disallow_untyped_defs"}},{"before":"a0b150a40c6e11f8d8be0b0a2ccf1ebb61499473","after":"cd8c2081d896c9f90f0a249a40a40214874d7394","ref":"refs/heads/gh/aorenste/56/head","pushedAt":"2024-06-12T15:06:43.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"aorenste","name":"Aaron Orenstein","path":"/aorenste","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/19294870?s=80&v=4"},"commit":{"message":"Update on \"Flip default value for mypy disallow_untyped_defs [final]\"\n\nNot requiring all functions to have types allows a lot of 'Any' types to slip in - which poison types and make mypy unable to properly typecheck the code.  I want to flip the default so that new files are required to have fully typed defs and we can have a burndown list of files that fail to require full types.\r\n\r\nThe preceding stack of PRs (cut up simply to limit the number of file changes per PR \"reasonable\") adds `# mypy: allow-untyped-defs` to any file which didn't immediately pass mypy with the flag flipped.  Due to changing files and merge conflicts it will probably be necessary to have several passes through before landing this final PR which turns the option on.\r\n\r\n\r\n\r\n\r\ncc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu fegin XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 mcarilli ptrblck leslie-fang-intel voznesenskym EikanWang Guobing-Chen zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire LucasLLC\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update on \"Flip default value for mypy disallow_untyped_defs [final]\""}},{"before":"c15856e00e669b507d1832683dd700f4df63fc72","after":"7138a632ff3d443ca8d0c00ba13275363ca037c5","ref":"refs/heads/gh/aorenste/56/base","pushedAt":"2024-06-12T15:06:41.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"aorenste","name":"Aaron Orenstein","path":"/aorenste","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/19294870?s=80&v=4"},"commit":{"message":"Update base for Update on \"Flip default value for mypy disallow_untyped_defs [final]\"\n\nNot requiring all functions to have types allows a lot of 'Any' types to slip in - which poison types and make mypy unable to properly typecheck the code.  I want to flip the default so that new files are required to have fully typed defs and we can have a burndown list of files that fail to require full types.\r\n\r\nThe preceding stack of PRs (cut up simply to limit the number of file changes per PR \"reasonable\") adds `# mypy: allow-untyped-defs` to any file which didn't immediately pass mypy with the flag flipped.  Due to changing files and merge conflicts it will probably be necessary to have several passes through before landing this final PR which turns the option on.\r\n\r\n\r\n\r\n\r\ncc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu fegin XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 mcarilli ptrblck leslie-fang-intel voznesenskym EikanWang Guobing-Chen zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire LucasLLC\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update base for Update on \"Flip default value for mypy disallow_untyp…"}},{"before":"2d8e2fae2faab28df36dda05fbff6ef61c316ba5","after":null,"ref":"refs/tags/ciflow/inductor/128239","pushedAt":"2024-06-12T15:03:49.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}},{"before":"2d8e2fae2faab28df36dda05fbff6ef61c316ba5","after":null,"ref":"refs/tags/ciflow/trunk/128239","pushedAt":"2024-06-12T15:03:47.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}},{"before":"4346129587f820e788f6a2f583cc799d137e030e","after":"c085d0184115921ee05873f40f79bfa983d41e80","ref":"refs/heads/gh/masnesral/75/orig","pushedAt":"2024-06-12T15:03:44.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"masnesral","name":null,"path":"/masnesral","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/135661981?s=80&v=4"},"commit":{"message":"[inductor] enable fx graph cache on torchbench\n\nSummary: We've already enabled for timm and huggingface, but we had failures saving cache entries for moco. It looks like https://github.com/pytorch/pytorch/pull/128052 has fixed that issue, so we can enable for torchbench.\n\nghstack-source-id: 4c4e9c90c7dbe5d9cb509693fa8216be75485fbf\nPull Request resolved: https://github.com/pytorch/pytorch/pull/128239","shortMessageHtmlLink":"[inductor] enable fx graph cache on torchbench"}},{"before":"2d8e2fae2faab28df36dda05fbff6ef61c316ba5","after":"abb48b233e6d061efb4978c8f7e620d04231bc93","ref":"refs/heads/gh/masnesral/75/head","pushedAt":"2024-06-12T15:03:41.000Z","pushType":"push","commitsCount":204,"pusher":{"login":"masnesral","name":null,"path":"/masnesral","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/135661981?s=80&v=4"},"commit":{"message":"Update on \"[inductor] enable fx graph cache on torchbench\"\n\n\nSummary: We've already enabled for timm and huggingface, but we had failures saving cache entries for moco. It looks like https://github.com/pytorch/pytorch/pull/128052 has fixed that issue, so we can enable for torchbench.\n\ncc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update on \"[inductor] enable fx graph cache on torchbench\""}},{"before":"82d7a36a27a26a4904f258fcdb79e37e91b510b6","after":"9ddd68ae2c59d55d68b278119b162635c14ab207","ref":"refs/heads/gh/masnesral/75/base","pushedAt":"2024-06-12T15:03:38.000Z","pushType":"push","commitsCount":203,"pusher":{"login":"masnesral","name":null,"path":"/masnesral","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/135661981?s=80&v=4"},"commit":{"message":"Update base for Update on \"[inductor] enable fx graph cache on torchbench\"\n\n\nSummary: We've already enabled for timm and huggingface, but we had failures saving cache entries for moco. It looks like https://github.com/pytorch/pytorch/pull/128052 has fixed that issue, so we can enable for torchbench.\n\ncc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 kadeng chauhang\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update base for Update on \"[inductor] enable fx graph cache on torchb…"}},{"before":"7baeea3a3d38e7a85bd84fb921b9530125897df8","after":null,"ref":"refs/tags/ciflow/inductor/127836","pushedAt":"2024-06-12T15:01:06.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}},{"before":"0625a85cc5764f2dd465b2c7d500102477d7b1f3","after":"46223e7e480132dc1881504b7687f79ec1abda1c","ref":"refs/heads/gh/aorenste/56/orig","pushedAt":"2024-06-12T15:01:01.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"aorenste","name":"Aaron Orenstein","path":"/aorenste","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/19294870?s=80&v=4"},"commit":{"message":"Flip default value for mypy disallow_untyped_defs\n\nghstack-source-id: 871717926b244416132a0b731c143932772ddd96\nPull Request resolved: https://github.com/pytorch/pytorch/pull/127836","shortMessageHtmlLink":"Flip default value for mypy disallow_untyped_defs"}},{"before":"7baeea3a3d38e7a85bd84fb921b9530125897df8","after":"a0b150a40c6e11f8d8be0b0a2ccf1ebb61499473","ref":"refs/heads/gh/aorenste/56/head","pushedAt":"2024-06-12T15:00:59.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"aorenste","name":"Aaron Orenstein","path":"/aorenste","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/19294870?s=80&v=4"},"commit":{"message":"Update on \"Flip default value for mypy disallow_untyped_defs [final]\"\n\nNot requiring all functions to have types allows a lot of 'Any' types to slip in - which poison types and make mypy unable to properly typecheck the code.  I want to flip the default so that new files are required to have fully typed defs and we can have a burndown list of files that fail to require full types.\r\n\r\nThe preceding stack of PRs (cut up simply to limit the number of file changes per PR \"reasonable\") adds `# mypy: allow-untyped-defs` to any file which didn't immediately pass mypy with the flag flipped.  Due to changing files and merge conflicts it will probably be necessary to have several passes through before landing this final PR which turns the option on.\r\n\r\n\r\n\r\n\r\ncc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu fegin XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 mcarilli ptrblck leslie-fang-intel voznesenskym EikanWang Guobing-Chen zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire LucasLLC\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update on \"Flip default value for mypy disallow_untyped_defs [final]\""}},{"before":"2804b7f261a9caede6abaab4aa7f372fe4a1efdc","after":"c15856e00e669b507d1832683dd700f4df63fc72","ref":"refs/heads/gh/aorenste/56/base","pushedAt":"2024-06-12T15:00:56.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"aorenste","name":"Aaron Orenstein","path":"/aorenste","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/19294870?s=80&v=4"},"commit":{"message":"Update base for Update on \"Flip default value for mypy disallow_untyped_defs [final]\"\n\nNot requiring all functions to have types allows a lot of 'Any' types to slip in - which poison types and make mypy unable to properly typecheck the code.  I want to flip the default so that new files are required to have fully typed defs and we can have a burndown list of files that fail to require full types.\r\n\r\nThe preceding stack of PRs (cut up simply to limit the number of file changes per PR \"reasonable\") adds `# mypy: allow-untyped-defs` to any file which didn't immediately pass mypy with the flag flipped.  Due to changing files and merge conflicts it will probably be necessary to have several passes through before landing this final PR which turns the option on.\r\n\r\n\r\n\r\n\r\ncc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu fegin XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 mcarilli ptrblck leslie-fang-intel voznesenskym EikanWang Guobing-Chen zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire LucasLLC\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update base for Update on \"Flip default value for mypy disallow_untyp…"}},{"before":"7d2ae0acdeb9db7041c6b6beee8cda876a8857ce","after":null,"ref":"refs/tags/ciflow/inductor/128453","pushedAt":"2024-06-12T14:56:36.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}},{"before":"f59c7f936bfe4488f8fce8611d69cffee4a21016","after":null,"ref":"refs/tags/ciflow/inductor/128428","pushedAt":"2024-06-12T14:56:36.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"pytorch-bot[bot]","name":null,"path":"/apps/pytorch-bot","primaryAvatarUrl":"https://avatars.githubusercontent.com/in/40112?s=80&v=4"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEY3gNlQA","startCursor":null,"endCursor":null}},"title":"Activity · pytorch/pytorch"}