hujunchao

Results 20 comments of hujunchao

Did you check accuracy of YOLOv4 by using your implementation on MSCOCO?

Hi,@ersheng-ai. Thank you for your great job! And I have a question! This project seems to only use yolov4' network and mosaic, but donot use Eliminate grid sensitivity，IoU threshold, CIOU...

Support dense detection like anchor-based YOLOV3/YOLOV4?

I have the same question.

LLaMA-13B (HF) Fails with OOM on a dual A100-80GB

There is a issue also about LLAMA-13B, https://github.com/tatsu-lab/stanford_alpaca/issues/93

通义模型

请问StructBERT 2.0现在开源了吗

[BUG]: torch.distributed.elastic.rendezvous.dynamic_rendezvous

WARNING:torch.distributed.elastic.rendezvous.dynamic_rendezvous:The node 'xxxxxx' has failed to send a keep-alive heartbeat to the rendezvous 'colossalai-default-job' due to an error of type RendezvousConnectionError.

[BUG] try to finetune a llama 33b on 8*A100 40G, 600G RAM. But always OOM on RAM.

same question

Fine-tuning

+1

Training LLaMa?

same question

High training loss of LLaMA 13B

It is a good question!

how to run on muti machine and muti GPUs？

same question