Skip to content

Latest commit

 

History

History
34 lines (20 loc) · 1.83 KB

developerDocs.md

File metadata and controls

34 lines (20 loc) · 1.83 KB

DSA AI Adapter Developer Documentation

Recommended Method to Build DSA AI Adapter

  1. Install DSA AI Adapter by following the instructions here.
  2. Deploy the model using conda as explained here.

Upon completion, Fast API will be initialized, and a URL for the hosted AI models will be generated.

Structure of the Fast API Root

Developers can access the AI model location by navigating to:

Location-of-DSA-ai-adapter/aiInferenceModel/

The code is organized as follows:

  • main.py: Executes the Fast API code, creating endpoints for each AI model.
  • utils.py: Contains instructions for downloading and pre-loading AI models.
  • ai_models folder: Houses individual code sets for pre-built AI models.
Segmentation Model

Developers can use the following code as a template for building custom segmentation models. After launching the Fast API, modifications to the code will be automatically reflected at the endpoint:

Classification Model

The following code serves as a template for building custom classification models: