The Deploy option in the Deploy & export page allows you to choose a model from the available experiments to choose as the corresponding AI assistant model.
In the example above, we have chosen MaskRCNN as the model we would like to deploy to the training assistant, as it gave the best results during model building. After this, if we go back to the annotation environment and choose assisted labelling, the Instance Segmentation AI assistant will be using the MaskRCNN model for masks.
Users can also use the deployed model for inference monitoring via API
Please note that deploying a model to a project via the Model Playground functionality makes it available for inference monitoring. So with the Hasty API and a selected model deployed, you can choose the specific model for inferencing.