Ice streams

use graph convolutional neural network for hyperspectral image classification

1

Hello, I find that you are also engaged in hyperspectral remote sensing image classification. Have you ever tried to use graph convolutional neural network for hyperspectral image classification? Can you...

Where is the Indian_best_3DCONV.hdf5

3

Hi,where is the file Indian_best_3DCONV.hdf5 the code you have writen in the Real3D_Conv.py best_weights_path = '/home/finoa/DL-on-HSI-Classification/Best_models/Indian_best_3DCONV.hdf5'

How to apply GCN to hyperspectral remote sensing image classification

1

Hello, I now have a 3d hyperspectral remote sensing image data cube, the pixel can be regarded as the graph node, and then each pixel can be labeled and classified....

hello，thanks to your work,How can I run the program? Will it work on Windows system?

FileNotFoundError: [Errno 2] No such file or directory: 'baseline-resnet50dilated-ppm_deepsup/encoder_epoch_20.pth'

7

Hello，thanks for your work. When I run train.py, error FileNotFoundError: [Errno 2] No such file or directory: 'Baseline - resnet50dilating-ppm_deepsup /encoder_epoch_20.pth'.Where can I find this file

ISPRS

2

hello,Could you share me with isprs's config file and eval ,thank you very much! my email [[email protected]](mailto:[email protected])

你好，我使用16G的RTX3080显卡设备试着跑了一下PU数据集和HongHu数据集，发现显存不够。究其原因，一方面是因为将HSI全影像输入，没有做batch_size处理，占用显存较大。另一方面是因为GAT和PAM处计算自注意力机制会占用较大的显存。请问你使用的是多大显存的RTX3090卡呢？

About source code

2

 Hi, in the BiSRNet.py, lines 149 and 154 do not correspond to the formulas (16)-(19) in the paper. Should it be changed to "energy1 = torch.bmm (q1, k1)" and...

Hi, Thanks for your work. I am very interested in your work and would like to reproduce it. But when I clicked on the download link for the preprocessed data,...

Thank you for your pioneering work, the current uploaded file seems to be missing the code related to self-supervised training, would you please provide the complete source code?