Create a model

A Model is a directory where you create, manage, and compare a set of Model Runs related to a machine learning task. A machine learning task is specified by the ontology of data that you use for Integration with model training service.

Labelbox is made up of an integrated suite of products that share a common structure of Data Rows, Labels, and Features. For example, in the diagram below, the global schema definition of Features is used in ontologies for labeling data as well as model training, testing, and evaluation in Models. Labelbox is structured so you can re-use feature schemas to create ontologies specific data labeling or model training tasks.

Create a Model via the UI

To create a new Model:

  1. Go to the Models tab and click + New Model.
  2. Configure the Model by giving it a name and a thumbnail.
  3. Select an ontology. The ontology determines which Data Rows can be used for model training. Once you select an ontology, all Data Rows containing the annotations within the selected ontology in your entire Catalog will be pulled in. You can further narrow the scope by selecting projects or datasets filter.
  4. Then, click Create model .

Create a Model via SDK

import labelbox 
from labelbox import Client

API_KEY = "YOUR_API_KEY"
client = Client(api_key=API_KEY)
PROJECT_ID = "YOUR_PROJECT_ID"
project = client.get_project(PROJECT_ID)

# create a model
lb_model = client.create_model(name=f"{project.name}-model",
                               ontology_id=project.ontology().uid)

What’s Next
Did this page help you?