search

jinnh / PDE-Net

[TIP2022] Deep Posterior Distribution-based Embedding for Hyperspectral Image Super-resolution
12 stars 1 forks source link

Questions about the result of testing the train model #2

Open Zhang9317112 opened 1 year ago

Zhang9317112 commented 1 year ago

In this code, I split the CAVE dataset into training dataset and testing dataset（20HS images for training,12HS images for testing）,processing them using the CAVE_data_processing.m you presented in the code.I trained the model and test it.The result was strange.In 3-th picture,PSNR is 80.9 and SSIM is 1.00!How could that be?I check the generated cave_train_test_filename.mat and it proved to split the dataset into training and testing dataset correctly.Please offer me some analysis,thank you.

tarxs commented 4 months ago

Hello, have you resolved this issue? Could you please let me know?