zhenqi-he / transnuseg

MICCAI2023 - TransNuSeg: A Lightweight Multi-Task Transformer for Nuclei Segmentation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

模型权重保存问题

sususu-lab opened this issue · comments

感谢您的论文,在训练阶段,模型的loss和对应图像可以存到日志当中,但模型训练的权重无法保存,显示This plugin does not support propagateSizeHints() This plugin does not support raise()想请教一下您有没有解决方法。同时请问只有训练文件 没有预测文件吗?

Hi, Sorry for the late response.
For the error related to saving model weights, could you please provide more details, like the detailed error information, python version, etc? And for the test document, you may find the dataset in this Google Drive (https://drive.google.com/drive/folders/1-ML_Z3yJOQsy3wbv__RL-qDZg7-n8-eI?usp=drive_link) and split it into train and test data , and in train.py line 289 - 309, there are some codes about setting model to evaluation mode and predict on the test data, they may be helpful to your question.