Adapting-OCR
Adapting-OCR copied to clipboard
error run train_semi_supervised.py
trafficstars
Hi please try to follow this blog, https://github.com/Deepayan137/Adapting-OCR/blob/master/notebooks/ocr.ipynb for the creation of a synthetic dataset. Also, if the current implementation gives you errors, you can try to modify the data pipeline so that it directly loads the train images and targets. Thanks
I modified a few codes and ran successfully. Thanks