tendra

Hi! I tried to solve the problem with the first error: ``` warnings.warn( github: fatal: ambiguous argument 'pose..origin/master': unknown revision or path not in the working tree. Use '--' to...

I try to modify `--device 0,1,2,3,4,5,6,7` into `--device 0` and set `OMP_NUM_THREADS` to `'1'`: `os.environ.setdefault(key_value, '1')` I get an error: ``` /usr/local/lib/python3.8/dist-packages/torch/distributed/launch.py:180: FutureWarning: The module torch.distributed.launch is deprecated and will...

I think I solve problems with error `labels require 56 columns each` by modifing `utils/datasets.py` Everything about the 17 key points has been changed. But I have an error again....

If I try to reduce --batch-size from 64 into 32: `!python train.py --data {data_location} --cfg {cfg_location} --weights {weights} --batch-size 32 --img 640 --kpt-label --project {my_project} --name edgeai-yolov5 --epochs 3 --hyp...

1. If I try not set OMP_NUM_THREAD:  I get same error `torch.cuda.OutOfMemoryError: CUDA out of memory.` 2. If I try to reduce memory load by reducing of --batch-size 64...

Everything about the 17 key points has been changed in utils/datasets.py.