12月1月2月3月4月5月6月7月8月9月10月11月
周一周三周五
一年内 259664 次贡献
更少更多
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 llmb-nemo-r2.4.0 分支比较 4 提交 »0a6b4439c7 Always use Nemotron tokenizer, in online or offline mode. (#15198)c47d82f433 Remove NCCL_NET_GDR_LEVEL and NCCL_NET_GDR_C2C environment variables. (#15162)66e17ac91f Set segment explicitly for nodes <= 18 (#15071)763ffa8b00 Support additional Slurm parameters (#14701)
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 yash/dev_llava_next 分支比较 10 提交 »9a588429ad finetune debug changese9d4b98131 Apply isort and black reformattingf6e92557c8 finetune changes397aa80f82 Apply isort and black reformatting590e7cd54e llava next end-end train
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 llmb-nemo-r2.3.0 分支4c7fb9a80c Add pyxis flags for writable and no-mount home. (#14385)
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 llmb-nemo 分支327e3bcfd7 Add profiling changes (#13484)
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 llmb-merge-2.3.0 分支355a7bfcc0 Merge origin/2.3.0 to llmb-merge-2.3.0
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 yash/chpk_llava_next 分支比较 6 提交 »cc8a95f5da added finetune recipe for nemo run88fbdac281 remove constant seq length padding52300d1ac9 strip left and right whitespace in sample encoder for VQA samples0e9241a600 change mock data module to support llava-next259850d450 add CI test for llava_next
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 llm_v2 分支比较 10 提交 »1370701e59 Apply isort and black reformatting43ec62fd20 Apply isort and black reformatting2bd41dd5e5 Remove old codefc7035d6fd Make nemo 1.x model loadable and usable21402de3a0 Make default save restore connector changeable
- frostbyte_neo 推送了仓库 github-featured/nvidia-nemo--speech 的 yangzhang/automodel_bert 分支比较 5 提交 »f5b084ab5a Apply isort and black reformatting7be794f145 adding BERT sequence classification functional testb6465dd735 fix some wrong class names7ded6f37bf Merge remote-tracking branch 'origin/main' into yangzhang/automodel_bert17c617fd5f sft sequence clasification single gpu for bert
