Error during new data set learning process

[yerin723]

hello.
Currently, we created a detection data set through python labelImg.py.
https://github.com/mindee/doctr/tree/main/references/detection
I also created a data format structure by referring to the address above.

But in the terminal window
python references/detection/train_pytorch.py ​​C:\Users\PIAI\Desktop\test C:\Users\PIAI\Desktop\val db_resnet50 --epochs 5 --device NVIDIA GeForce RTX 2080

When I type it like this
ImportError: cannot import name 'VOCABS' from partially initialized module 'doctr.datasets' (most likely due to a circular import) (C:\Users\PIAI\Desktop\ocrproject\doctr\doctr\datasets_init_.py)

This error keeps occurring.

What should I do to resolve the error?
I am a beginner, so please explain it step by step.

[felixT2K]

Hi @yerin723 👋🏼

1. Fork the repo and clone it
2. cd doctr -> pip3 install -e .[torch]
3. pip3 install fastprogress wandb psutil
4. python3 references/detection/train_pytorch.py <path/to/your/train_set> <path/to/your/val_set> db_resnet50 --epochs 5 --device 0