Open sangeun99 opened 1 year ago
Hello Im trying to fine-tune your nice model with LRS2 datasets
I tried to do this with single GPU but i faced this error
With your answer here #22, I tried again with 4 GPUs but it doesn't work without any logs.
I didn't modify any codes but only the path for the files. Could you check what is wrong with my options? This is my sbatch file to run the code.
` #!/bin/bash
singularity exec --cleanenv -H $PWD:/home --nv tensorflow.sif /bin/bash 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 12 \ --kp_num 15 \ --generator DepthAwareGenerator tensorboard --logdir log/DaGAN/log
exit 0 `
Thanks in advance!
Hello Im trying to fine-tune your nice model with LRS2 datasets
I tried to do this with single GPU but i faced this error
With your answer here #22, I tried again with 4 GPUs but it doesn't work without any logs.
I didn't modify any codes but only the path for the files. Could you check what is wrong with my options? This is my sbatch file to run the code.
` #!/bin/bash
SBATCH --job-name=rundata
SBATCH --gres=gpu:4
SBATCH --cpus-per-gpu=8
SBATCH --mem-per-gpu=10G
SBATCH --time 1-0
SBATCH --partition batch_ce_ugrad
SBATCH -o slurm/logs/slurm-%A.out
singularity exec --cleanenv -H $PWD:/home --nv tensorflow.sif /bin/bash 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 12 \ --kp_num 15 \ --generator DepthAwareGenerator tensorboard --logdir log/DaGAN/log
exit 0 `
Thanks in advance!