{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":651693080,"defaultBranch":"main","name":"3DCoMPaT-challenge","ownerLogin":"Cattalyya","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2023-06-09T20:44:40.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/15096930?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1687371679.0","currentOid":""},"activityList":{"items":[{"before":"bf0219e188f9a8fce9fc17e3d03859199af94f09","after":"2aa9197d6b095cbbd67f240246a693f992f94320","ref":"refs/heads/main","pushedAt":"2023-06-22T18:59:51.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"README: Update overview, performance and how-to train in the main README.","shortMessageHtmlLink":"README: Update overview, performance and how-to train in the main REA…"}},{"before":"f0957bed9cdaba4a74e15b0f3696d067a21460cc","after":"bf0219e188f9a8fce9fc17e3d03859199af94f09","ref":"refs/heads/main","pushedAt":"2023-06-22T18:58:42.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"README: Update overview, performance and how-to train in the main README.","shortMessageHtmlLink":"README: Update overview, performance and how-to train in the main REA…"}},{"before":"5719543a2f1bf7acb554c5ea47ba02a1c2cbb967","after":"f0957bed9cdaba4a74e15b0f3696d067a21460cc","ref":"refs/heads/main","pushedAt":"2023-06-22T18:57:28.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"README: Update overview, performance and how-to train in the main README.","shortMessageHtmlLink":"README: Update overview, performance and how-to train in the main REA…"}},{"before":"15b9d0e2d1fcc3d5912f2b42ca54cdc6b4e8ae7e","after":"5719543a2f1bf7acb554c5ea47ba02a1c2cbb967","ref":"refs/heads/main","pushedAt":"2023-06-22T18:55:03.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"README: Update overview, performance and how-to train in the main README.","shortMessageHtmlLink":"README: Update overview, performance and how-to train in the main REA…"}},{"before":"9062bc3e2cdc4a3f5f630b02aa767d222c80f2b1","after":"15b9d0e2d1fcc3d5912f2b42ca54cdc6b4e8ae7e","ref":"refs/heads/main","pushedAt":"2023-06-22T18:51:57.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"README: Update overview, performance and how-to train in the main README.","shortMessageHtmlLink":"README: Update overview, performance and how-to train in the main REA…"}},{"before":null,"after":"11ecc1ec6688062870aec19d9796f70e362940a0","ref":"refs/heads/rgb","pushedAt":"2023-06-21T18:21:19.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"2D_3D: Add segformer flag and add fine model dir.\n\nOnly enable segformer 2D seg if asked for it - not used in predict.\nnit: Update to latest cls model","shortMessageHtmlLink":"2D_3D: Add segformer flag and add fine model dir."}},{"before":null,"after":"e47a757815309d79cb1f0741e428a4c9bc5c0a97","ref":"refs/heads/cls6d","pushedAt":"2023-06-21T18:19:49.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"[exp] try cls 6D","shortMessageHtmlLink":"[exp] try cls 6D"}},{"before":null,"after":"35b58c340ff322da89d6e2e34ecdbe0c5d585f5a","ref":"refs/heads/partmat","pushedAt":"2023-06-21T18:19:31.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"3D: partmatseg: Training group seg (partmat) together.","shortMessageHtmlLink":"3D: partmatseg: Training group seg (partmat) together."}},{"before":null,"after":"63134e34373d29671d783515dc593ac9dd0ef83e","ref":"refs/heads/features","pushedAt":"2023-06-21T18:19:13.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"3D: 2D_3D loader is slow, so predict3D loads from 3D loader and read pre-inferenced 2D features\n\n2D segmentation logits were saved to hdf5 file, so we can now read that value directly without re-run 2D inference again.\nNow this commit switch to use 3D loader to read 2D features from files instead of using 2D_3D loader,\nwhich takes long time due to 2D having 8x dataset size compared to 3D and mapped to 8 of them mapped\nto the same 3D models, so using 3D loader speed up by 8 times.","shortMessageHtmlLink":"3D: 2D_3D loader is slow, so predict3D loads from 3D loader and read …"}},{"before":null,"after":"e09b15429d5e8fc50a5465be3d08a26e98a99084","ref":"refs/heads/cls","pushedAt":"2023-06-21T18:18:41.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"2D: cls: Add EfficientNet for 2D cls prediction\n\nStill has issue. Training works but reloading model for eval got wrong accuracy.","shortMessageHtmlLink":"2D: cls: Add EfficientNet for 2D cls prediction"}},{"before":"aa6b1205399b2483c0b50fbfa42ddb68936e4753","after":"9062bc3e2cdc4a3f5f630b02aa767d222c80f2b1","ref":"refs/heads/main","pushedAt":"2023-06-21T18:09:36.000Z","pushType":"push","commitsCount":37,"pusher":{"login":"Cattalyya","name":"Cattalyya","path":"/Cattalyya","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/15096930?s=80&v=4"},"commit":{"message":"Merge branch 'main' into mergergb","shortMessageHtmlLink":"Merge branch 'main' into mergergb"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAADRyeVOgA","startCursor":null,"endCursor":null}},"title":"Activity · Cattalyya/3DCoMPaT-challenge"}