Giters

harlanhong / CVPR2022-DaGAN

Official code for CVPR2022 paper: Depth-Aware Generative Adversarial Network for Talking Head Video Generation

Home Page:https://harlanhong.github.io/publications/dagan.html

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

There seems to be a problem with distributed running codeWhen I entered the training command, the program did not respond

pfeducode opened this issue · comments

commented

When I entered the training command, the program did not respond. How can I solve it
11

commented

Hi, @pfeducode! Did you solve this problem?

I also met this problem, but I couldn't solve this problem.

If you solve this problem, will you share the any idea?

commented

I deleted the distributed code, and then it can run normally。

commented

There seems to be a problem with distributed running code. When I entered the training command, the program did not respond, and I had to delete the distributed code

commented

Please use this command line:
CUDA_VISIBLE_DEVICES=0,1,2,3 python run_dataparallel.py --config config/vox-adv-256.yaml --device_ids 0,1,2,3 --name DaGAN_voxceleb2_depth --rgbd --batchsize 48 --kp_num 15 --generator DepthAwareGenerator

commented

Actually, I also met this problem when I was using another version pytorch. It seems to only work for "1.9.0+cu111".
image

commented

Actually, I also met this problem when I was using another version pytorch. It seems to only work for "1.9.0+cu111". image

Okay, I'll try later

commented

After removing the distributed code for the generator and discriminator and making device changes in the "model_dataparallel.py" file, I have successfully got it working on a single GPU.