{"full_name":"tensorflow/tensor2tensor","owner":"tensorflow","name":"tensor2tensor","shard":"1","url":"https://github.com/tensorflow/tensor2tensor","description":"Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.","homepage":"","topics":["machine-learning","machine-translation","deep-learning","reinforcement-learning","tpu"],"tag_pages":["deep-learning","machine-learning","machine-translation","reinforcement-learning"],"language":"Python","license":"Apache-2.0","avatar":"https://avatars.githubusercontent.com/u/15658638?v=4","created_at":"2017-06-15T16:57:39Z","pushed_at":"2023-06-02T18:55:09Z","metrics":{"stars":17471,"forks":3829,"watchers":9,"issues":1245,"prs":671,"releases":75,"issues_open":574,"issues_closed":671,"prs_open":15,"prs_closed":150,"prs_merged":506,"discussions":0},"growth":{"day":{"stars":0,"forks":0,"base_day":"2026-11-15"}},"history":{"dates":["2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[17472,17472,17473,17473,17471,17471,17471],"forks":[3808,3819,3830,3831,3831,3829,3829],"watchers":[null,9,9,9,9,9,9],"issues_open":[null,574,574,574,574,574,574],"prs_open":[null,15,15,15,15,15,15]},"boards":{},"same_owner":[{"full_name":"tensorflow/tensorflow","name":"tensorflow","shard":"1","boards":{"total":33}},{"full_name":"tensorflow/models","name":"models","shard":"1","boards":{}},{"full_name":"tensorflow/tfjs","name":"tfjs","shard":"1","boards":{}}],"issues":[{"number":444,"title":"[Tuning] Results are GPU-number and batch-size dependent","comments":81,"created_at":"2017-11-27T08:07:40Z"},{"number":317,"title":"Unable to reproduce WMT En2De results","comments":52,"created_at":"2017-09-21T21:29:35Z"},{"number":1643,"title":"Stuck after printing 'Successfully opened dynamic library libcublas.so.10.0'","comments":45,"created_at":"2019-07-24T16:59:00Z"},{"number":975,"title":"distributed training fail","comments":29,"created_at":"2018-08-07T07:00:01Z"},{"number":529,"title":"Bug introduced in v1.3.0 causing training divergence","comments":27,"created_at":"2018-01-19T18:26:10Z"},{"number":621,"title":"Inference speed for Transformer Vs GNMT. Transformer is slower","comments":23,"created_at":"2018-02-27T04:08:18Z"},{"number":1221,"title":"[Feature Request] Mixed Precision","comments":22,"created_at":"2018-11-13T16:51:42Z"},{"number":415,"title":"How to measure the number of training epochs","comments":22,"created_at":"2017-11-13T16:45:04Z"}],"discussions":[],"latest_release":{"name":"v1.15.7","tag":"v1.15.7","published_at":"2020-06-17T16:10:01Z","body":" * Multistep Adam Optimizer many thanks to @AgoloCuongHoang for contributing in #1773 !\r\n * Residual Shuffle-Exchange Network thanks to @EmilsOzolins in #1805 !\r\n * Not pinning the gym version.","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}