Train a Model

Start a training job from the Python SDK.

You can start a training job either in the Roboflow platform or using the Python SDK, using the instructions below.

Python SDK

To schedule a training job via the Python SDK, use the train() method. You can train on a version that does not have an associated model, or you can train at the project level. If you train at the project level, a new dataset version will be created then a training job will start.

Note: this method will finish when the Roboflow platform has started a training job, not when the training job is finished.

import roboflow

rf = roboflow.Roboflow(api_key=YOUR_API_KEY_HERE)

# List all projects for your workspace
workspace = rf.workspace()

# get a project
project = rf.workspace().project("PROJECT_ID")

# create a new version and train
new_version = project.generate_version()
version = project.version(new_version)

# train on an existing version
# (the one we created above)
version.train()

# train on a new version
project.train()