leonselina

Results 8 comments of leonselina

error when compiling DCNv2

it works, thank you:)

src/deform_conv_cuda_kernel.cu(258): error: no suitable constructor exists to convert from "c10::ScalarType" to "at::Type"

Did you fix the error? I got the same problem.

为什么不接国内的大模型

文新不太适合coding....but

Can not train llama-7b due to OOM on 40GA100

A6000x2 with 48GB VRAM, got OOM too.

[Chatllama]: MultiGPU support for training

I added accelerate in the code as #233 ,bug got error: Traceback (most recent call last): File "/nvmessd0/nebullvm/apps/accelerate/chatllama/artifacts/main.py", line 3, in from chatllama.rlhf.actor import ActorTrainer File "/home/spzq/.local/lib/python3.10/site-packages/chatllama/rlhf/actor.py", line 17, in...

[Chatllama]: MultiGPU support for training

Hi @PierpaoloSorbellini , I trained Llama 7B with deepspeed, but got error: "MP=1 but world size is 2". How can I train Llama 7B with multi-GPU? because the limits of...

[Chatllama] Errors when training actor model based on LLaMA-7B

I got the error too! The WORLD_SIZE can not be get

测试刚好选择的是A股上升的时间段

基于RL的agent交易，和传统的量化交易是不同的概念，没有所谓“未来函数”吧！