search

ying09 / TextFuseNet

A PyTorch implementation of "TextFuseNet: Scene Text Detection with Richer Fused Features".
MIT License
475 stars 122 forks source link

TEXTFUSENET_MUTIL_PATH_FUSE_ON #69

Closed hossein-moghaddam closed 3 years ago

hossein-moghaddam commented 3 years ago

Hi. Could anyone tell me what is this "TEXTFUSENET_MUTIL_PATH_FUSE_ON" in the configs doing? because when this is enabled I get this error: The size of tensor a (2) must match the size of tensor b (36) at non-singleton dimension 0 but when I disable this the error is gone. but I'm afraid this could affect the performance. thank you so much.

Real-YeJ commented 3 years ago

@h311-git "TEXTFUSENET_MUTIL_PATH_FUSE_ON" means using the multi_path_features_fusion module.

hossein-moghaddam commented 3 years ago

@h311-git "TEXTFUSENET_MUTIL_PATH_FUSE_ON" means using the multi_path_features_fusion module.

Thank you so much for your response. but do you have any idea why am i getting this error when this is enabled?

hossein-moghaddam commented 3 years ago

I found the solution :). the problem was because I had changed the batch size. anyway, thank you.