Inference on test folder

[amandhapola]

Hi,
I was trying to infer using the best trained SVTRv2 model using this
command : !python3 OpenOCR/tools/infer_rec.py \ --c ./models/svtrv2_smtr_gtc_ch/config.yml \ --o Global.infer_img="../dataset/icdar_dataset/rec_dataset/test" \ --o Global.pretrained_model="./output/rec/svtrv2_lnconv_smtr_gtc_ch/best.pth"

but this gives output :

Config after merge: None [2025/03/20 19:37:53] openrec INFO: finetune from checkpoint ./output/rec/svtrv2_lnconv_smtr_gtc_ch/best.pth Infering images from None Traceback (most recent call last): File "OpenOCR/tools/infer_rec.py", line 385, in <module> main(cfg.cfg) File "OpenOCR/tools/infer_rec.py", line 345, in main for file in get_image_file_list(cfg['Global']['infer_img']): File "/paddle/openocr/OpenOCR/tools/utils/utility.py", line 37, in get_image_file_list raise Exception("not found any img file in {}".format(img_file)) Exception: not found any img file in None

. Basically the infer_img config is not being passed .

My settings :
ubuntu20.04
python 3.8.10