{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":768971032,"defaultBranch":"main","name":"ao","ownerLogin":"melvinebenezer","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2024-03-08T04:29:21.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/6395936?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1726922856.0","currentOid":""},"activityList":{"items":[{"before":null,"after":"c5c36f4da711b64c16515d5df156f548d5253181","ref":"refs/heads/uintx-ops","pushedAt":"2024-09-21T12:47:36.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"fix get_plain in UintxAQTLayout\nadd new test cases for ops in Uintx","shortMessageHtmlLink":"fix get_plain in UintxAQTLayout"}},{"before":"8aa6533ae08d97664105b03d0e93bcbacd50da0b","after":"0bdde92114b470823aa24725bf3b0811e980c8ce","ref":"refs/heads/main","pushedAt":"2024-09-21T06:20:53.000Z","pushType":"push","commitsCount":12,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"Rename Floating point to fp8 (#909)","shortMessageHtmlLink":"Rename Floating point to fp8 (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2539248022\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/909\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/909/hovercard\" href=\"https://github.com/pytorch/ao/pull/909\">pytorch#909</a>)"}},{"before":"ba2d3b1333b90ccd0186216649a1c58c6a17ce56","after":"8aa6533ae08d97664105b03d0e93bcbacd50da0b","ref":"refs/heads/main","pushedAt":"2024-09-16T08:37:42.000Z","pushType":"push","commitsCount":59,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"Switch back to supported CI runners (#887)","shortMessageHtmlLink":"Switch back to supported CI runners (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2525766921\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/887\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/887/hovercard\" href=\"https://github.com/pytorch/ao/pull/887\">pytorch#887</a>)"}},{"before":"ba1e09024dc1d906319364c713a48c91103e70d5","after":"ab5ffd71a234652d6c67268fbbaaa9024d8254e4","ref":"refs/heads/int4_refactor","pushedAt":"2024-08-30T16:00:35.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"fix: adapt to implements from utils","shortMessageHtmlLink":"fix: adapt to implements from utils"}},{"before":"42170df563607b3b1e82ff24c066b0d9a070e59b","after":"ba1e09024dc1d906319364c713a48c91103e70d5","ref":"refs/heads/int4_refactor","pushedAt":"2024-08-30T05:21:22.000Z","pushType":"push","commitsCount":140,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"Merge branch 'main' into int4_refactor","shortMessageHtmlLink":"Merge branch 'main' into int4_refactor"}},{"before":"e5df48e1033611d6433dc3fd1de7923e0f555ebc","after":"ba2d3b1333b90ccd0186216649a1c58c6a17ce56","ref":"refs/heads/main","pushedAt":"2024-08-30T05:11:34.000Z","pushType":"push","commitsCount":124,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"add in formatting skip everything (#779)\n\nghstack-source-id: 34fe56595eac4d1a2fecb07a230307c0b2b767d7\r\nPull Request resolved: https://github.com/pytorch/ao/pull/688","shortMessageHtmlLink":"add in formatting skip everything (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2495182870\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/779\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/779/hovercard\" href=\"https://github.com/pytorch/ao/pull/779\">pytorch#779</a>)"}},{"before":"891a5884d0bccc00787c4e0d94f8812df2439604","after":"e5df48e1033611d6433dc3fd1de7923e0f555ebc","ref":"refs/heads/main","pushedAt":"2024-07-25T03:54:25.000Z","pushType":"push","commitsCount":7,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"[test] Fix regression tests (#537)\n\n* Fix regression tests\r\n\r\nSummary:\r\n\r\nTorch 2.2 is compiled with numpy 1.x, but when we `pip install -r\r\nrequirements-dev.txt` we download a higher version of numpy (2.0)\r\n\r\nThis causes an error with the .numpy() calls and importing torch in\r\ngeneral. I don't think we want to pin the versions in\r\nrequirements-dev.txt, so instead I added a pin to the numpy version in\r\nthe specific torch spec, so it'll only run for 2.2.\r\n\r\nPT 2.3+ support numpy 2.0+ which is why those test don't fail.\r\n\r\nTest Plan:\r\n\r\nReviewers:\r\n\r\nSubscribers:\r\n\r\nTasks:\r\n\r\nTags:\r\n\r\n* use == and specify for CPU as well\r\n\r\n* update","shortMessageHtmlLink":"[test] Fix regression tests (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2428148267\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/537\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/537/hovercard\" href=\"https://github.com/pytorch/ao/pull/537\">pytorch#537</a>)"}},{"before":"56d46a2869d5342568e3d746d4b907966e0341ce","after":"891a5884d0bccc00787c4e0d94f8812df2439604","ref":"refs/heads/main","pushedAt":"2024-07-18T01:04:34.000Z","pushType":"push","commitsCount":22,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"Add doc for serialization/deserialization of torchao optimized models (#524)\n\nSummary:\r\nAddressing following questions:\r\n\r\n1. What happens if I save a quantized model\r\n2. What happens if I load a quantized model and describing deteails like assign=True\r\n\r\nSpecifically\r\n1. Do you need ao as a dependency when you're loading a quantized model\r\n2. Is the saved quantized model smaller on disk than the unquantized one\r\n\r\nTest Plan:\r\n.\r\n\r\nReviewers:\r\n\r\nSubscribers:\r\n\r\nTasks:\r\n\r\nTags:","shortMessageHtmlLink":"Add doc for serialization/deserialization of torchao optimized models (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2414655728\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/524\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/524/hovercard\" href=\"https://github.com/pytorch/ao/pull/524\">…</a>"}},{"before":"d54399c1dd3238dde36325505d75de0676314a9d","after":"42170df563607b3b1e82ff24c066b0d9a070e59b","ref":"refs/heads/int4_refactor","pushedAt":"2024-07-17T00:50:54.000Z","pushType":"push","commitsCount":35,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"Merge branch 'main' into int4_refactor","shortMessageHtmlLink":"Merge branch 'main' into int4_refactor"}},{"before":"c2f9b84604536a72804787001c1b63daae792ee9","after":"56d46a2869d5342568e3d746d4b907966e0341ce","ref":"refs/heads/main","pushedAt":"2024-07-07T06:10:42.000Z","pushType":"push","commitsCount":20,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"Update low-bit Adam benchmark (#481)\n\n* update benchmark\r\n\r\n* add rank1 option to lpmm\r\n\r\n* add comma\r\n\r\n* update readme\r\n\r\n* remove unwanted file\r\n\r\n* update","shortMessageHtmlLink":"Update low-bit Adam benchmark (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2393514590\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/481\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/481/hovercard\" href=\"https://github.com/pytorch/ao/pull/481\">pytorch#481</a>)"}},{"before":"c0497b826aef17ed59f87806645dccd4c8e04622","after":"d54399c1dd3238dde36325505d75de0676314a9d","ref":"refs/heads/int4_refactor","pushedAt":"2024-07-07T04:56:20.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"updated uint4 and perchannel_symmetricweight based on new API #391","shortMessageHtmlLink":"updated uint4 and perchannel_symmetricweight based on new API <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2358541926\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/391\" data-hovercard-type=\"issue\" data-hovercard-url=\"/pytorch/ao/issues/391/hovercard\" href=\"https://github.com/pytorch/ao/issues/391\">pytorch…</a>"}},{"before":"a686e6b907bb85001fda3ad1023263cec22b3529","after":"c0497b826aef17ed59f87806645dccd4c8e04622","ref":"refs/heads/int4_refactor","pushedAt":"2024-06-28T05:28:54.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"refactored per channel symmetric weight","shortMessageHtmlLink":"refactored per channel symmetric weight"}},{"before":null,"after":"a686e6b907bb85001fda3ad1023263cec22b3529","ref":"refs/heads/int4_refactor","pushedAt":"2024-06-28T05:07:31.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"refactor uint4 to implements pattern","shortMessageHtmlLink":"refactor uint4 to implements pattern"}},{"before":"4ca3985be603e6496da7ec57adf1942c8b32a78e","after":"c2f9b84604536a72804787001c1b63daae792ee9","ref":"refs/heads/main","pushedAt":"2024-06-28T00:31:26.000Z","pushType":"push","commitsCount":100,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"ghstack flow test - 3 (#443)\n\nSummary:\natt\n\nTest Plan:\ntesting flow\n\nReviewers:\n\nSubscribers:\n\nTasks:\n\nTags:\nPull Request resolved: https://github.com/pytorch/ao/pull/443\nApproved by: https://github.com/clee2000, https://github.com/huydhn","shortMessageHtmlLink":"ghstack flow test - 3 (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2375950477\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/443\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/443/hovercard\" href=\"https://github.com/pytorch/ao/pull/443\">pytorch#443</a>)"}},{"before":"3c9bc20d7d5f7d33ee9790d03a2f7dc558693645","after":"4ca3985be603e6496da7ec57adf1942c8b32a78e","ref":"refs/heads/main","pushedAt":"2024-05-26T05:21:15.000Z","pushType":"push","commitsCount":13,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"Improve primitives for FP6 quant (#248)","shortMessageHtmlLink":"Improve primitives for FP6 quant (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2299593183\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/248\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/248/hovercard\" href=\"https://github.com/pytorch/ao/pull/248\">pytorch#248</a>)"}},{"before":"e9e5fae34de1071d39b03fbde97b7f0565b1d02b","after":"3c9bc20d7d5f7d33ee9790d03a2f7dc558693645","ref":"refs/heads/main","pushedAt":"2024-05-21T03:00:13.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"ci: Add push trigger for binary build workflows (#259)","shortMessageHtmlLink":"ci: Add push trigger for binary build workflows (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2306878686\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/259\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/259/hovercard\" href=\"https://github.com/pytorch/ao/pull/259\">pytorch#259</a>)"}},{"before":"55e5d40bf7d8fa99efd6c75d647f462197159852","after":"e9e5fae34de1071d39b03fbde97b7f0565b1d02b","ref":"refs/heads/main","pushedAt":"2024-05-20T16:10:57.000Z","pushType":"push","commitsCount":144,"pusher":{"login":"melvinebenezer","name":"James Melvin Ebenezer","path":"/melvinebenezer","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6395936?s=80&v=4"},"commit":{"message":"Fix quant_primitive dtype that caused perf regression (#253)\n\nSummary:\r\nAPI call in quantize_activation_per_token_absmax is not exactly preserving the original\r\ncode, this PR fixes that\r\n\r\nTest Plan:\r\nneed to check perf with torchbenchmarks\r\n\r\nReviewers:\r\n\r\nSubscribers:\r\n\r\nTasks:\r\n\r\nTags:","shortMessageHtmlLink":"Fix quant_primitive dtype that caused perf regression (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2303410235\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/ao/issues/253\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/ao/pull/253/hovercard\" href=\"https://github.com/pytorch/ao/pull/253\">pytorch#253</a>)"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"Y3Vyc29yOnYyOpK7MjAyNC0wOS0yMVQxMjo0NzozNi4wMDAwMDBazwAAAAS8s-6m","startCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wOS0yMVQxMjo0NzozNi4wMDAwMDBazwAAAAS8s-6m","endCursor":"Y3Vyc29yOnYyOpK7MjAyNC0wNS0yMFQxNjoxMDo1Ny4wMDAwMDBazwAAAARO-W4w"}},"title":"Activity · melvinebenezer/ao"}