Training Multiple Images with Shared Style and Individual Context Prompts

[star4s]

I am interested in training the DreamStyler model using five images that share a similar artistic style but have distinct content. For each image, I would like to provide a unique context prompt that highlights specific details or elements in the image. Is there a way to train the model with these five images simultaneously, using a shared style reference while incorporating individual context prompts for each image? If so, could you please provide guidance on how to structure the input files or modify the training script to accommodate this setup? The goal is to fine-tune the model to generate new images that capture both the shared style and the unique context of each input image.

Thank you!