Hsu Wan Ting

OSError: [Errno 24] Too many open files during training

4

@MhLiao I tried to pretrain a model on SynthText dataset. I followed your training script: ``` python -m torch.distributed.launch --nproc_per_node=4 tools/train_net.py --config-file configs/pretrain/seg_rec_poly_fuse_feature.yaml ``` But I always got the OSError...

When I tried to convert ICDAR2013, I got the error below: ``` FileNotFoundError: [Errno 2] No such file or directory: '../datasets/icdar2013/test_gts/img_1.jpg.txt' ``` I found that there are no `test_gts` in...

Accuracy gap between single GPU and multiple GPUs

3

I'm using lm-eval v0.4.2 to evaluate Llama 7b on the open llm leaderboard benchmark. I found that there are accuracy gaps between single GPU and multiple GPUs as below. (I...