search

Qsingle / LearnablePromptSAM

Try to use the SAM-ViT as the backbone to create the learnable prompt for semantic segmentation
Apache License 2.0
84 stars 14 forks source link

Excuse me, which images are to be trained for --image and --mask_path? #17

Open zys1999 opened 3 months ago

zys1999 commented 3 months ago

Hello, author. I am trying to train the AROI dataset using your model. I set the paths for the original images and label images as --image AROIonline/inter_intra/raw and --mask_path AROIonline/inter_intra/groundTruth, respectively. I trained for 32 epochs with a learning rate of 0.05, but the IoU was only 0.016 after training. Could you please advise on how to correctly set the paths for the original images and label images? Thank you.

zys1999 commented 3 months ago

Additionally, I have not processed the AROI dataset, and I set numClass to 256. What other steps are necessary to achieve approximately 40% as mentioned in your paper?

Qsingle commented 3 months ago

Thank you for your question. You can set the --mask_path for the AROI dataset to patient/mask/number. For example, patient12/mask/number and set the --image to patient12/raw/labelled.