Train CRNN with semi-supervised methods Pseudo Label (PL) miss "saved_model"

ku21fan / STR-Fewer-Labels

Scene Text Recognition (STR) methods trained with fewer real labels (CVPR 2021)

MIT License

173 stars 27 forks source link

Train CRNN with semi-supervised methods Pseudo Label (PL) miss "saved_model" #14

Closed ecnuycxie closed 1 year ago

ecnuycxie commented 1 year ago

The step "Train CRNN with semi-supervised methods Pseudo Label (PL)" is incomplete as it does not include the required argument "saved_model".

ku21fan commented 1 year ago

Hi,

The step "Train CRNN with semi-supervised methods Pseudo Label (PL)" does not require the argument "saved_model" Instead, it uses "--model_for_PseudoLabel" For example,

CUDA_VISIBLE_DEVICES=0 python3 train.py --model_name CRNN --exp_name CRNN_PL --Aug Crop90-Rot15 \
--semi Pseudo --model_for_PseudoLabel saved_models/CRNN_aug/best_score.pth

Please check the README.md

Best, Jeonghun