ying09 / TextFuseNet

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

RuntimeError:CUDA out of memory #46

Open watanka opened 3 years ago

watanka commented 3 years ago

I'm aware that this is more hardware issue of mine, but I was wondering if there is any way I can make the model little bit smaller to save GPU memory. Thank you in advance!

Real-YeJ commented 3 years ago

@watanka use smaller image sizes

bhrt-sharma commented 3 years ago

is there any other solution, to this problem?

25b3nk commented 3 years ago

As suggested by @Real-YeJ , after reducing the MIN_SIZE_TEST parameter and MAX_SIZE_TEST parameter in ./configs/ocr/icdar2015_101_FPN.yaml, I could get the demo running on my system. One can change the corresponding config file for the demo they want to run.