Florian Zimmermeister
Florian Zimmermeister
https://huggingface.co/spaces/BlinkDL/RWKV-World-7B Demo https://huggingface.co/models?other=rwkv Models I will check other PRs and try to make an integration
@Narsil interested ?
https://github.com/LAION-AI/Open-Assistant/issues/3232#issuecomment-1694762888 @andreaskoepf
Same here Most of my answers won't be accept because I already answered. No account switch or something else
For cases like realtime ASR more optimized methods, for example as rust modules, would be super cool. Maybe with functionality for community pipelines as in diffusers, just for speech ?
Pinging @ver217 who made the integration
with manually half precision data this occurs: ``` ?it/s]Traceback (most recent call last): File "/mnt/nvme0/asr/run_speech_recognition_ctc.py", line 360, in main() File "/mnt/nvme0/asr/run_speech_recognition_ctc.py", line 353, in main trainer.fit(model=plmodel, train_dataloaders=dloader) File "/home/aware/anaconda3/lib/python3.9/site-packages/lightning/pytorch/trainer/trainer.py", line...
Thanks to project like hivemind and other community members like @fxtentacle it is possible to organize the computing capacity. If Hajo is still interested in that use case we could...
Will definitely take a look next week Great to see it merged now 💪
If you implement it or link some useful code for training we could provide some computing power