matciotola / lambda-pnn Goto Github PK

Hello @matciotola
Thank you for your great work!
I get interested in your code and I wanted to run it but I have a question about the dataset that you used for training. Is it free and accessible data for the public? if yes, how can I achieve this dataset? if not, do you have any suggestions on what other free datasets are available and similar to the one you have used?

Hello,

I have been following your method to process some images in TIFF format. However, when I attempted to test it, I encountered an issue. The error occurred in the input_propocessing.py file, specifically at the line where it attempts to concatenate two tensors: img_in = torch.cat((ms_exp, pan), dim=1) / max_value.

The sizes of the ms_exp and pan tensors are as follows:

ms_exp: torch.Size([1, 3, 1024, 1024])
pan: torch.Size([1, 1, 3, 1024, 1024])
As I am new to Deep Learning, I’m having trouble understanding why this error is occurring and how to resolve it. Could you please provide some guidance or suggestions to help me troubleshoot this issue? I would greatly appreciate your assistance.

Thank you very much.

in Usage part, "!python tiff_mat_conversion.py -m Tiff2Mat -ms /path/to/ms.tif -pan /path/to/ms.tif -o path/to/file.mat" must be changed to "!python tiff_mat_conversion.py -m Tiff2Mat -ms /path/to/ms.tif -pan /path/to/pan.tif -o path/to/file.mat"