SpA-GAN_for_cloud_removal
SpA-GAN_for_cloud_removal copied to clipboard
Cloud Removal for High-resolution Remote Sensing Imagery based on Generative Adversarial Networks.

I use the pre-trained model to test in RICE1, and calculate the PSNR and SSIM. I get the results as follows, not the same to the paper: Avg. PSNR: 28.8865...
Hi, can I get train/test split list for RICE 2, like RICE 1 you uploaded ? Thanks
from eval.py in the line 2 cannot import name 'compare_ssim' from 'skimage.measure'
hello,can you display the format of "predict.py"? usage: predict.py [-h] --config CONFIG [--test_dir TEST_DIR] --out_dir OUT_DIR --pretrained PRETRAINED [--cuda] [--gpu_ids GPU_IDS] [--manualSeed MANUALSEED] predict.py: error: argument --config: expected one argument
I'm trying to run the Code with CPU so I changed the map_location=None to map_location=torch.device('cpu') . Now I become the TypeError: Device(): argument 'type' (position 1) must be str, not...
The pre-training model provided works fine, but I still want to ask how do I save the training model with "train.py"? Looking forward to your anwser, thank you!