Yushun Xiang
Yushun Xiang
Actually, I think chage `has_img = 'images' in samples.keys()` to `has_img = samples['data_type'] == 'multi'` is better and more elegant.
I have the same problem. I use Tesla V100 GPUs (each 32GB memory)
```shell torchrun --nproc-per-node NUM_GPU train.py --cfg-path train_configs/minigptv2_finetune.yaml ```
> > ```shell > > torchrun --nproc-per-node NUM_GPU train.py --cfg-path train_configs/minigptv2_finetune.yaml > > ``` > > I suppose --nproc-per-node specifies the number of GPUs on a machine, right? Do you...
I have the same question.