change_detection.pytorch
change_detection.pytorch copied to clipboard
In the final step, “infer_vis” can't work.
In the verification step of the last step, each “tiff” cannot be opened, which causes the program to end directly and cannot view the accuracy. I guess it's the "infer_vis" problem, have you ever encountered this situation?
Could you describe your problem in more detail? e.g. some code or logs
Below is the display of the error report for my runtime program。After ensuring no bugs in the front, this should be a read error that occurred in the last validation function. I don't know how to fix it to get the program to run properly. " Model saved! valid: 76%|███████▌ | 97/128 [00:25<00:07, 3.90it/s]TIFFOpen: ./res\test_1.tif: Cannot open. TIFFOpen: ./res\test_10.tif: Cannot open. TIFFOpen: ./res\test_100.tif: Cannot open. TIFFOpen: ./res\test_101.tif: Cannot open. TIFFOpen: ./res\test_102.tif: Cannot open. TIFFOpen: ./res\test_103.tif: Cannot open. TIFFOpen: ./res\test_104.tif: Cannot open. TIFFOpen: ./res\test_105.tif: Cannot open. TIFFOpen: ./res\test_106.tif: Cannot open. "
Does the 'res' folder exist? If it doesn't exist, make it manually
There's actually no "image reading". Maybe it's a platform (window?) problem?
I'm sorry that it's not convenient for me to debug the code now.
If it still doesn't work, tell me and I'll fix it later. thx
hhhhh,You are right.Thank u .