Seyyed Ali Ayati comments

Results 15 comments of


                                            Seyyed Ali Ayati

Lacking post conditions for "make field final"

Refactoring file --> refactorings/make_field_final.py

[sui tools] prototype of Sui surfer testing tool

@sblackshear >>> It was able to call about 18% of the functions. How can we do better? I am interested... and ideas to get start?

torch.cuda.OutOfMemoryError: CUDA out of memory When Trying to Save the Model

> https://github.com/MHketbi/starcoder > > try my fork Worked on NVIDIA A100 80 GB, but not on NVIDIA A40 48 GB

Is there any code for preparation of the dataset on which Starcoder has been originally trained?

>I've been successfully able to finetune Starcoder on my own code May I know your hardware that you used for finetuning?

During finetuning (finetuner_starcoder.py), I'm running out of GPU memory during the checkpoint saving step (save_pretrained)

Facing same issue with 4 A40 48 GB

Cannot search for string prefix because * gets escaped

I have also the same issue. Is there any way for partial search?

Feature Request: Get object's raw source

This is indeed a very useful feature. I started working on it.

Feature Request: Get object's raw source

> This is indeed a very useful feature. I started working on it. Here is my solution: 1. Set `loc=True` in parse function. 2. Use the following helper function: ```python...

ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: -9) local_rank: 0

> I don't know if that has a link. But I was using cuda 11.6, but it showed me warnings that I had to use version 1.17. And since also...

ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: -9) local_rank: 0

> maybe because of CPU OOM ? Can you provide more details please?