Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
suryatmodulus pushed 69 commits to main suryatmodulus/refiners
- ella adapter implementation. tested with sd1.5 model 277b0fd
- ella: refresh README and mkdocs Follow up of #389 428c2dd
- chore(licence-update): Update PyProject Toml - License b6f547d
- add comfyui custom nodes b67efb2
- add comfyui registry github action 76b4994
- add box segmenter comfyui examples workflow 78ec8cf
- add comfyui registry badge dc1a2e1
- nits: cross-reference the box-segmenter a061f00
- add comfyui requirements.txt c2e255c
- bump comfyui-refiners version 1cee7c6
- fix comfyui requirements.txt bdcc0aa
- bump comfyui-refiners version cf247a1
- ⭐ Add example code for Stable Diffusion(1.5) (#409) 5aef140
- add karras sigmas to dpm solver af6c5ae
- MultiUpscaler: improve resizing logic - Do not have a parameter to downscale (caller can do it beforehand if they wa... a51d695
- update `gaussian_blur` fluxion util, see https://github.com/pytorch/vision/commit/45e053b2ae470cc3ad69c3ac9c378eeff41... 2395e66
- loosen the tolerances for some sam tests (because of a recent pytorch upgrade) 6d58492
- implement _add_noise for dpm solver c6631c7
- move some tests into the adapters test folder 444882a
- add missing device and dtype to SD1UNet's UpBlocks 602c63c
- and 49 more ...
suryatmodulus pushed 1461 commits to main suryatmodulus/NeMo
- Change FIM Dataset Random Seed Init (#9165) * change seed to dataset init * Apply isort and black reformatting ... 7a23bfa
- increase time limit for Speech_Checkpoints_tests (#9186) 43686ec
- fix ep rank (#9161) Signed-off-by: Gao Deng <[email protected]> 467d94b
- TRTLLM new API support (#9003) * Add trtllm checkpoint * Change model config * fix no query_group * Using b... 77090d4
- Alit/optim 8k (#9166) * fix fuser issue with dynamo * optimized 4k seq len * optim 8k * add checkpointing ... b1628cf
- Bucketing duration bins: less optimal but instant init when not provided + fixes in estimation script (#9157) * Buck... 93907f0
- Enable CUDA graphs by default only for transcription (#9196) (#9197) * Enable CUDA graphs only for transcription. Sy... acbd4e0
- move tts fixtures (#9183) * move tts fixtures Signed-off-by: Jason <[email protected]> * Apply isort and black... 4167641
- enable matryoshka embedding learning (#9130) * enable matryoshka embedding learning Signed-off-by: arendu <adithy... 4d574fe
- Add guards to SD imports (#9158) * Add guards to SD imports Signed-off-by: yaoyu-33 <[email protected]> * [p... 5df8e11
- Implement async distributed checkpoint save (#9028) * Prevent duplicated checkpoints Signed-off-by: Mikołaj Błaż ... c2daa91
- Fix incorrect checkpoint removal logic (#9192) * Fix incorrect if logic Signed-off-by: Mikołaj Błaż <mblaz@nvidia... 1de4b49
- Update to using Model Optimizer (formerly AMMO) in PTQ workflow (#9178) * Update PTQ to use nvidia-modelopt Signe... 6cb618a
- GPU-based vectorized Specaug Version 2 (#9155) * GPU-based vectorized SpecAug Signed-off-by: Piotr Żelasko <petez... 061cc45
- run_cicd_for_release_branches_also (#9213) 964ea3c
- Update nemo.export module for quantized models (#9218) * Remove config aligner - no longer needed after TRT-LLM 0.9 ... d0a4535
- Update index.rst (#9080) Removed best-practices.rst file Signed-off-by: jgerh <[email protected]... b489fba
- ci: Speeding NeMo-CI up by using caching (#9174) * build: Add `Dockerfile.ci` Signed-off-by: Oliver Koenig <okoen... 526b6ad
- Add save option to the TRT-LLM export test script (#9221) * Add save option to the test script Signed-off-by: Onu... e465b9c
- rename paths2audiofiles to audio (#9209) (#9220) * rename paths2audiofiles to audio * update transcribe to au... 73edac4
- and 1441 more ...
suryatmodulus pushed 108 commits to master suryatmodulus/VoiceCraft
- gradio app added 5cef625
- deprecated .update not used anymore, better error handling, can use voicecraft without whisper 74fa659
- global models f9fed26
- add Dockerfile 1a1390e
- update start jupyter script d7bd722
- bugfixes, seed support, better ui 1a219cf
- whisperx added 3d3f32b
- modify the Dockerfile, download correct lib versions 2506954
- Merge branch 'jasonppy:master' into master bbe3437
- add Colab to readme 40887bd
- install whisperx deps 97b1f51
- fix whisperx loading issue, update generation instruction 0d19fa5
- README update, gradio_app.ipynb update, debug print removed 94e9f9b
- Merge branch 'master' of github.com:zuev-stepan/VoiceCraft 871ae70
- smart transcript fix 6f71fa6
- configurable tmp path 21b69ad
- Update README.md 1141775
- Adding share=True b9ffda9
- Update gradio_app.py 5c92b28
- Update gradio_app.py fcdd1d3
- and 88 more ...
suryatmodulus pushed 3 commits to main suryatmodulus/fish-speech
- Add details about how to start agent. (#651) * Update Start_Agent.md * Remove livekit * Update doc * [pre-c... 7b0802d
- fix the bug where non-english speech will always prepend sentences with 'speak' (#652) ec2c5b7
- [docs]:Add docs about fish agent. (#654) * [docs]Add docs of Fish Agent. * [docs]:Fix some issues * [pre-commi... aaca85b
suryatmodulus pushed 161 commits to main suryatmodulus/moondream
- add cuda support 2fb9571
- Merge pull request #22 from spartanhaden/main add cuda support 15d5dd7
- Refactor code and remove unused imports ddf4a06
- Refactor text_model.py and vision_encoder.py 7a4d6c3
- Refactor PhiConfig class in configuration_phi.py d5b825d
- Refactor InferenceParams and Embedding classes 3226f9f
- Refactor rotary embedding function in modeling_phi.py 32382c2
- Refactor _apply_rotary_emb_kv function to improve readability and performance 873b392
- Refactor rotary embedding calculation in modeling_phi.py 025c8d4
- Refactor rotary embedding functions 085ee6b
- Refactor rotary embedding functions in modeling_phi.py 89dde7d
- Refactor RotaryEmbedding class to use non-trainable buffers 1be1147
- Update import statements and remove unused code fc71d33
- Refactor RotaryEmbedding class to improve precision and performance 3f0b043
- Refactor forward method in RotaryEmbedding class bcdfaa6
- Refactor MLP module in modeling_phi.py 107f9e3
- Refactor SelfAttention class in modeling_phi.py 66427ec
- Add Flash Attention module 30fc465
- Add Flash Attention module 0607a68
- Refactor forward method in CrossAttention class 3b9598e
- and 141 more ...
suryatmodulus pushed 4 commits to main suryatmodulus/together-cookbook
suryatmodulus pushed 31 commits to main suryatmodulus/Amphion
- fix the bug for valle resume for two stages a7b7bf9
- update VALLE recipe 74de5ab
- Fix typos and improve clarity in README 4ad5e5f
- Merge branch 'main' into valle_resume b5eca77
- Update tts_trainer.py bd3ce83
- Merge pull request #144 from lmxue/valle_resume Fix bug for VALLE resume 1e4a650
- Fix bugs for content extractor's args (#165) 5cb75d8
- Remove useless file (#180) d335514
- VALL-E: new version release (#220) A refined version of VALL-E. We have changed the underlying implementation to Lla... f96a153
- Update TTS readme (#224) 5dfe9fd
- Release Emilia and Emilia-Pipe (#227) ba335d2
- Fix Emilia README.md typo (#237) * Fix Emilia README.md typo * Update README.md * Update README.md 65ddb1e
- fix parameters for fs2 (#225) f00c3ac
- Improve the code of WER (#210) efcdf4b
- Add Jets implementation (#231) 72112a6
- Fix VALLE inference bug (#253) a17f139
- Training & Inference code for FAcodec (#229) * Training & Inference code for FAcodec * Update vocoder_trainer.py *... 211e1d4
- Update SingVisio Code (#264) * Update SingVisio Code * fix: black format 8dce8b3
- Emilia is now publicly available! (#265) * Update README.md --------- Co-authored-by: Yuan_Tuo <yuantuo666@gma... b4aecf8
- Emilia is available on HuggingFace (#267) * Update README.md --------- Co-authored-by: Yuan_Tuo <yuantuo666@g... 05a1e7e
- and 11 more ...