mahrokh3409
commented
8 months ago Dear @HamedAlemo, I finetuned the multi_temporal_crop_classification based on the instructions in GitHub, and the resulting model is "best_mIoU_epoch_80.pth".
I wanted to run the inference using the provided example in the Github as below:
python model_inference.py -config /home/abd037/IBM_HLS/hls-foundation-os/configs/multi_temporal_crop_classification.py -ckpt /home/abd037/IBM_HLS/Mah_crop/best_mIoU_epoch_80.pth -input /home/abd037/Downloads/images -output /home/abd037/Downloads/Mah -input_type tif -bands 0 1 2 3 4 5
The inference ran successfully, but the resulting tiff image is black.
May I know if the config file should be the one I used for fine-tuning? I used the same config file in the above command. would appreciate it if you could help me with this issue.
Kind Regards, Mah
Dear @hanxLi,
Happy New Year.
I finetuned the multi_temporal_crop_classification based on the instructions in GitHub, and the resulting model is "best_mIoU_epoch_80.pth".
I wanted to run the inference using the provided example in the Github as below:
python model_inference.py -config /home/abd037/IBM_HLS/hls-foundation-os/configs/multi_temporal_crop_classification.py -ckpt /home/abd037/IBM_HLS/Mah_crop/best_mIoU_epoch_80.pth -input /home/abd037/Downloads/images -output /home/abd037/Downloads/Mah -input_type tif -bands 0 1 2 3 4 5
The inference ran successfully, but the resulting tiff image is black.
May I know if the config file should be the one I used for fine-tuning? I used the same config file in the above command. I would appreciate it if you could help me with this issue.
Kind Regards, Mah