Post Model Training

This guide details how to use your newly trained model, which is now in your local model cache. You must add your new model to a project, either locally or via publishing to the model catalog. Then you can see how your model performs by installing the model and starting up your application.

Add the Model to a Project

There are two ways to add a model to a project: Locally or via the Dashboard.

Add a Model to a Project Locally

  • Navigate to the project directory from a terminal window

  • Run the add models command, using the --local-version flag to add the model to alwaysai.app.json

$ aai app models add <username/modelname> --local-version <version>

Add the Model to a Project via the Dashboard

Publish the Model to the Model Catalog

  • Run publish command to add the model to “My Models” on your Dashboard

$ aai model publish username/modelname

Add the Model to an Existing Project via the Dashboard

Once you have successfully uploaded your model to the model catalog, you can add it to an existing project.

  • Navigate to the that project in the ‘Projects’ tab in the Dashboard, and viewing the models section

image

  • Then, you can click on the ‘+’ symbol to choose from your models

image

Use the Model in the Project

Modify app.py

  • Change the model used in app.py by replacing alwaysai/modelname in main() at the top of the file to use your desired model, like so:

image

Start the Application

  • Install the model, using aai app install if on a Windows or Mac, or aai app build if on a Linux or deploying to the edge. You can always visit this page for information on building and starting applicatins.

$ aai app install

OR

$ aai app build
  • Run the Project

Run the project using the aai app start command

$ aai app start