kaveh8/wsi-ddpm repository overview

⁠WSI Generation with DDPM

⁠Overview

This Docker image allows you to generate synthetic Whole Slide Imaging (WSI) patches using the Diffusion-based Denoising Probabilistic Model (DDPM). By running this image, you will have access to a user-friendly Gradio interface hosted on localhost:8080.

⁠GPU

If you prefer to run the inference on a GPU, follow these steps:

1. Install the Nvidia Toolkit to utilize the host Nvidia drivers:

   sudo apt install nvidia-container-toolkit
   sudo nvidia-ctk runtime configure
   

   Copy

2. Pull the Docker image:

   sudo docker pull kaveh8/wsi-ddpm:latest
   

   Copy

3. Run the Docker container using the following command:

   sudo docker run --rm -p8080:8080 -it --gpus all kaveh8/wsi-ddpm
   

   Copy

⁠CPU

If a GPU is not available, you can still run the image on a CPU, although it may take longer to process.

1. Pull the Docker image:

   sudo docker pull kaveh8/wsi-ddpm:latest
   

   Copy

2. Run the Docker container using the following command:

   sudo docker run --rm -p8080:8080 -it kaveh8/wsi-ddpm
   

   Copy

---

After running the image, open your web browser and navigate to the localhost:8080 to start generating samples.

Feel free to customize the Docker command based on your requirements and system configuration.

Enjoy generating synthetic WSI patches effortlessly with this Docker image!