- 14 Jul, 2022 1 commit
-
-
Yanghan Wang authored
Summary: Pull Request resolved: https://github.com/facebookresearch/d2go/pull/332 Solve: "we have decoupled qat with `.training` in quantization, maybe we should use some flags in `cfg` instead of checking this attribute here as well" Reviewed By: jerryzh168 Differential Revision: D37801241 fbshipit-source-id: ed9884d7b462da195ed2e07c42634acfe5beefb2
-
- 27 Jun, 2022 1 commit
-
-
Mik Vyatskov authored
Summary: Pull Request resolved: https://github.com/facebookresearch/d2go/pull/315 This method is no longer needed given the there's a better alternative: save/load_binary_output. Reviewed By: tglik, wat3rBro Differential Revision: D37307668 fbshipit-source-id: 598a6d7aeb68f65fa6b42a230175c37d19fd7798
-
- 18 Jun, 2022 1 commit
-
-
Tsahi Glik authored
Summary: Pull Request resolved: https://github.com/facebookresearch/d2go/pull/297 X-link: https://github.com/facebookresearch/mobile-vision/pull/84 Add command line arg to specify whether and where to save results. This is useful where binaries are being launched from another process, or remotely on another machine. Reviewed By: wat3rBro Differential Revision: D37157955 fbshipit-source-id: 2a48cf967f6cf928049f2be41952834e1dd2a04d
-
- 15 Jun, 2022 1 commit
-
-
Mik Vyatskov authored
Summary: Pull Request resolved: https://github.com/facebookresearch/d2go/pull/289 Right now configs are written to a dedicated folder after training, one file per model config. This PR introduces a new function that allows to read model configs in the same format, for the situations where the configs cannot be passed back directly, e.g. when running through torchx. Reviewed By: wat3rBro Differential Revision: D37086940 fbshipit-source-id: 3938381bcf48a8069fb4b840fd2c2d052e983c6c
-
- 16 May, 2022 1 commit
-
-
Miquel Jubert Hermoso authored
Summary: Pull Request resolved: https://github.com/facebookresearch/d2go/pull/223 Reviewed By: wat3rBro Differential Revision: D35928340 fbshipit-source-id: de740101e8bed9d1df84bbed438ea3611d2eb015
-
- 23 Jun, 2021 1 commit
-
-
Yanghan Wang authored
Summary: Pull Request resolved: https://github.com/facebookresearch/d2go/pull/90 Reviewed By: zhanghang1989 Differential Revision: D29279123 fbshipit-source-id: d94cea65bd439d54fd14afded0dba066799cedca
-
- 16 Jun, 2021 1 commit
-
-
Kai Zhang authored
Summary: This diff logs D2 (https://github.com/facebookresearch/d2go/commit/692a4fb3c506aeebbb49070a20d139d617381b19)Go model instantiation events to table scuba_caffe2_pytorch_usage_stats, so that we could track model usage in fblearner, bento, local scripts, etc. Reviewed By: zhanghang1989 Differential Revision: D28986723 fbshipit-source-id: 3e865354e5884c9e82bd1b08819cc10d349f93bd
-
- 08 Mar, 2021 1 commit
-
-
Yanghan Wang authored
Summary: PathManager is the old version, therefore config files are not saved to Manifold Reviewed By: sstsai-adl Differential Revision: D26882435 fbshipit-source-id: 089c53db90080737b10bafe43d154d34221a7ce2
-
- 03 Mar, 2021 2 commits
-
-
Kai Zhang authored
Summary: As titled. Make a copy of quantization callback to unblock D2go OSS. Reviewed By: zhanghang1989 Differential Revision: D26735525 fbshipit-source-id: 12b77f04cfa1361e856b26ea218a262da1fadd88
-
facebook-github-bot authored
fbshipit-source-id: f4a8ba78691d8cf46e003ef0bd2e95f170932778
-