Susnato Dhar comments

Results 46 comments of


                                            Susnato Dhar

low fps and no bounding box on Tensorflow 2

Hi @sachinkmohan , can you please provide some additional information about how to reproduce this issue? Any Colab notebook or something would be useful. Actually I created TF 2.6 version...

Filter on dataset too much slowww

Hi @lhoestq, any updates on this issue? The `filter` method is still veryyy slow 😕

Filter on dataset too much slowww

Hi @lhoestq, thanks a lot for the update! I would like to work on this(if possible). Could you please give me some steps regarding how should I approach this? Also...

Ernie-M

Hi, @shermansiu is there any pytorch/tf implementation of this model?

Ernie-M

@shermansiu Thanks for the resources! I am currently trying to port the model to huggingface(pytorch), (done till Embedding Layer with acceptable tolerance of 1e-3)

Ernie-M

Hi @KnutJaegersberg, Ernie-M is implemented!

Community contribution: Adding Flash Attention 2 support for more architectures

I would like to work on `OPT`.

Community contribution: Adding Flash Attention 2 support for more architectures

@younesbelkada yes I have.

Community contribution: Adding Flash Attention 2 support for more architectures

Since @sahilbhosale63 is not working on ` GPTBigCode (Starcoder)`(as he said [here](https://github.com/huggingface/transformers/issues/26350#issuecomment-1733685691)) can I take that @younesbelkada?

Community contribution: Adding Flash Attention 2 support for more architectures

Hi @younesbelkada , I am talking this up for `GPT-neo`.