ying09 / TextFuseNet

A PyTorch implementation of "TextFuseNet: Scene Text Detection with Richer Fused Features".
MIT License
476 stars 123 forks source link

corrupted pre-trained models #98

Closed dinalzein closed 2 years ago

dinalzein commented 2 years ago

Hello, I'm trying to run the demo icdar2013_detection.py with pytorch=1.7.1 and cuda=11.0; I was able to do the step-by-step installation.txt and build the 'TextFuseNet-detectron2' using with that version but I encountered the following error when using the pre-tained model model_syn_r101_pretrain.pth or the fine-tuned model model_ic15_r101.pth which I got from here :

RuntimeError: unexpected EOF, expected 11233569 more bytes. The file might be corrupted.