Creating an Interface for Your Custom Algorithm

Bilayers allows you to easily create a web interface (using Gradio) or a Jupyter Notebook interface for any deep-learning cell segmentation algorithm by filling out a simple configuration file in YAML format. It’s as simple as filling out a Google Form! Before you begin, let’s break down the key elements you need to know -

1. Interface
   • What is Gradio?
   • What is JupyterNB?
2. Understanding config.yaml requirements
   • parameters
   • display_only
   • results
   • exec_function
   • docker_image
   • algorithm_folder_name
   • citations
3. Steps to create your custom Algorithm’s Interfaces
   • Step 0: Clone the repo
   • Step 1: Create your own Algorithm spec-file
   • Step 2: Use the custom Docker Images
   • Step 3: Raise the PR → Bilayers Team will approve and can publish the docker-image on docker-hub