drimeF0

Results 9 comments of drimeF0

Accurate report?

![image](https://github.com/bloomberg/memray/assets/67423319/b392e840-add3-40e2-a666-0f54b7940c5a) Same problem

RuntimeError: Failed to import transformers.models.llama.modeling_llama

same problem

Error training GPT-2 with LoRA INT8

> > Thank you! That resolved the issue with the devices. > > However, now the following error appears: > > ``` > > [/usr/local/lib/python3.8/dist-packages/peft/tuners/lora.py]) in forward(self, x) > >...

Video Reader's get_metadata function fails on videos with sound

![image](https://github.com/pytorch/vision/assets/67423319/be8de32c-5969-461f-b969-9d7a6b2e8003) same

Requirements?

like regular stable diffusion

RuntimeError: weight should have at least three dimensions

same error, my code was supposed to train vae on 8 v2 TPU using fsdp ![image](https://github.com/huggingface/accelerate/assets/67423319/70423f73-5388-49a1-8d1c-c98f0ad8647c)

training loop freezes after first step on TPU

![image](https://github.com/huggingface/accelerate/assets/67423319/a1c5e771-eea7-4fd0-b964-e5e7de399e5c) without accelerate

training loop freezes after first step on TPU

first step takes 48~ seconds

Flow Matching Euler Scheduler added

> Well It is not useable with StableCascadeCombinedPipeline as is. This this scheduler should be used to train new models. With this scheduler diffusion models tend to converge faster, due...