Tanmay Laud

icon indicating a website link www.linkedin.com/in/tanmaylaud

icon location California icon location Research @hippocraticaidev Past: Research at @megagonlabs ,@icebergnlp, Lead ML @omdena. Contributed to @streamlit, @huggingface, @deepset-ai

Results 20 comments of Tanmay Laud

Make utilities available for Jupyter Users too

@koaning I am guessing Rasa uses some form of beam search. In that case a minimum span tree could be created and then the chosen beam could be highlighted. And...

Converting the course to an audiobook

Thanks for sharing @yemikifouly !

Create a dataset loader for ShAReCLEF 2014 Task 2

#self-assign

Create dataset loader for BioASQ Task Synergy

#self-assign

Create a dataset loader for CLEF eHealth 2019, Task 1

#self-assign

ParaNMT: sentencepiece model and .pt model vocab and embedding size mismatch?

I am facing the same issue

What is the expected behaviour with changing the bandwidth parameter

@jishengpeng awaiting your response on this. Thanks!

[Bug] Unable to fix model output

having this issue with 70b bf16, 405b bf16 and fp8. is there a root cause analysis on this ?

[Bug] Unable to fix model output

It is the flashinfer sampling that is causing the determinism issue. Do we have any fixes for that? Torch sampling with argmax is deterministic but it is slower.

Alignment language vocabulary and speech space

@jishengpeng what did the training data look like for the text-speech alignment ?