Running code on single GPU

[MLRadfys]

Hi and thanks for you're great work and for making the code public!

I tried to run fully supervised training on a single GPU, but unfortunately the validation values are very low and both training and validation loss are not changing (about the same value). I started training using the following command:

python -m torch.distributed.launch --nproc_per_node=1 --master_port 28887 --use_env train_clip_vg.py --num_workers 4 --epochs 110 --batch_size 64 --lr 0.00025 --lr_scheduler cosine --aug_crop --aug_scale --aug_translate --imsize 224 --max_query_len 77 --sup_type full --dataset unc+ --data_root /path_to_image_data --split_root /path_to_split --output_dir /path_to_output/unc+;

Now Iam wondering if Iam doing something wrong?

Thanks in advance for any advice,

best regards,

M