Hi, I want to use Stable Diffusion for my custom MRI images with 4 channels input (size: 4,256,256), my dataset includes 2 types: "tumor" or "non-tumor" images. I'm wondering whether it is better to fine-tune the pre-trained model or train it from the scratch?
In case of fine-tuning, is there any code for the direction?
Thanks in advance!
Hi, I want to use Stable Diffusion for my custom MRI images with 4 channels input (size: 4,256,256), my dataset includes 2 types: "tumor" or "non-tumor" images. I'm wondering whether it is better to fine-tune the pre-trained model or train it from the scratch? In case of fine-tuning, is there any code for the direction? Thanks in advance!