11月12月1月2月3月4月5月6月7月8月9月10月
周一周三周五
一年内 258841 次贡献
更少更多
- frostbyte_neo 推送了仓库 github-featured/rasahq--rasa 的 multiprocessing-queue-update 分支4c9da289fc Update multiprocessing queue and pass it as an argument instead
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 gpt_core_transformer_dist_ckpt 分支比较 10 提交 »cd8e6b2faa add distributed checkpointing for interleaved6941fdb03a add distributed checkpointing for optimizer states92660002f0 add get_module_list86e322eed8 revert self._model, fix device after load, add prefixd4f62eb623 working on load in progress
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 r1.20.0_pt_23.09_sharp 分支a8e364187e slym/sharp
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 chtruong/r2.3.0-cherry-pick-pt8 分支比较 2 提交 »d63265d3d0 Fix handling Llama Embedding dimensions param and prompt type in the ONNX export tutorial (#13262)af77fb9eb0 Add Llama4 Scout and Maverick Support (#12898)
- frostbyte_neo 推送了仓库 github-featured/rasahq--rasa 的 moodbot-w-rules 分支85d7026e8e Merge branch 'master' into moodbot-w-rules
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 chtruong/r2.3.0-cherry-pick-pt7 分支比较 10 提交 »3221afeb64 Revert "Append current PR to build cache"68f9d8300c Reopen 13040 (#13199)c74f728ee8 Revert "build: `MAMBA_TAG=2e16fc3062cdcd4ebef27a9aa4442676e1c7edf4` (#13173)"c2c67749c9 Revert "Fix build cache"dd7c61706c Fix build cache
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 r1.20.0_pt_23.09 分支比较 6 提交 »bd0b818ac2 Broadcast loss only when using pipeline parallelism and within the pipeline parallel domain (#7576)26d5e112eb Hotfix (#7501)165c275adb Cache FP8 weight and transpose only at the first micro-batch in each validation and test routine (#7470)c592d4099b Add an option to disable manual GC in validation (#7467)18701b0ac5 Cherry pick PR #7003 (#7441)
