search

wenwenyu / PICK-pytorch

Code for the paper "PICK: Processing Key Information Extraction from Documents using Improved Graph Learning-Convolutional Networks" (ICPR 2020)
https://arxiv.org/abs/2004.07464
MIT License
553 stars 191 forks source link

How do I remove the part of the image segment? #104

Open dhwgithub opened 2 years ago

dhwgithub commented 2 years ago

When I remove the image segment during the encoder phase, I'll be prompted for the following error when training： image

However, when only 【images=self.cnn(images) 】 is not annotated, the program will train normally.

How do I modify the code when I completely remove the image segment?（The code can train properly before it is removed）