CVC-ClinicDB数据集上test.py，结果差距比较大

wurenkai / H-vmunet

[arXiv] The official code for "H-vmunet: High-order Vision Mamba UNet for Medical Image Segmentation".

89 stars 3 forks source link

CVC-ClinicDB数据集上test.py，结果差距比较大 #3

Closed LWTvickyzz closed 1 month ago

LWTvickyzz commented 7 months ago

3090显卡，resume_model下载的，跑出来的结果差距比较大，请问是什么原因呢？ Quicker_20240407_111756

wurenkai commented 7 months ago

Hi, we just confirmed that the weight files are correct. We have uploaded a total of three weight files for three different datasets, did you modify the correct corresponding weight files when testing the CVC-ClinicDB dataset. Also, you can check if there are any formatting issues with the data preparation by following the 'Prepare your own dataset' steps. I hope my answer can help you. test_CVC

wurenkai commented 7 months ago

Also check that the image size is correctly resized to 256*256 when your data is prepared.

Richard1Bob commented 1 month ago

Hi, we just confirmed that the weight files are correct. We have uploaded a total of three weight files for three different datasets, did you modify the correct corresponding weight files when testing the CVC-ClinicDB dataset. Also, you can check if there are any formatting issues with the data preparation by following the 'Prepare your own dataset' steps. I hope my answer can help you.

你训练了多久呢