ayews / m3net Goto Github PK

您好通过你们给的M3Net-R的权重进行评估，但是得到的结果和你们论文提供的不一致，但是你们用你们提供的预测图进行评估得到的结果是一致的，所以想问下是哪里有问题，显卡和你们用的是一样的3090

The current link does not seem to be working. Is there an alternate link (google drive)?

Dear author:

I would like to express my sincere gratitude for sharing your code and paper openly.

I have a specific question regarding the pre-training weights of ResNet50 and Swin-Transformer. Were the pre-training weights of these models trained on ImageNet1k? I have observed some discrepancies between ResNet50 and the implementation in torchvision.

Thank you for your time and consideration. I greatly appreciate your contribution to the research community and look forward to your response.

Hello author, I have used the prediction map provided by your model in my paper to compare the picture effect, but the journal requires the license to use the relevant pictures, and the author's electronic signature or institutional signature is needed, I have sent the email to you or the corresponding author of the paper, can you reply to me, thank you very much.

Hi, congratulations for your great work achieving state-of-the-art performances on several benchmarks.

However, I found that our source code was used in your repository without any mentioning on ReadMe page and there's no Licence for the source code. Our source code is under the MIT Licence, so you need to include our Licence for the part of your source code which have been brought from ours.

Yours - dataloader.py | custom_transforms.py
Ours - dataloader.py | custom_transforms.py

Keep up your good work and please understand other people's hard work.