Getting started
Overview
We created this Python API so you can access all the functionalities of the Labelbox API without having to write GraphQL queries. This documentation will guide you through installation, setup, and some common use cases.
See the Python API repo in Github.
pip install --upgrade labelbox to upgrade (please consult the Changelog before upgrading). The Python API supports Python versions 3.6 and 3.7.Python API features
- Read, create, and modify objects and object-to-object relationships
- Simplify your data import with bulk Data Row creation
- Download a JSON export of labels
- Prioritize, re-enqueue & adjust the count of labels in the labeling queue
- Use webhooks with limitations
- Support for
PredictionandPredictionModeldata types
Installation & authentication
Step 1 Create your API key
All users wishing to access the Python API must have an API key for authentication. For instructions on creating a key, see API keys.
Step 2 Install
Install Labelbox using pip.
user@machine:~$ pip install labelbox
Step 3 Authenticate
Pass your API key as an environment variable. Then, import and initialize the API Client.
user@machine:~$ export LABELBOX_API_KEY="<your_api_key_here>"
user@machine:~$ python3
from labelbox import Client
client = Client()
Passing in a custom endpoint is only applicable for on-premises use cases. If this applies to you, you may pass the API key and server endpoint explicitly when you initialize the Client object. Otherwise, refer to the step above.
from labelbox import Client
client = Client("<your_api_key_here>", "https://your-server.com/labelbox/")
Python script
Here is a Python snippet you can also use to connect to the API. Run this as a python script and make sure to include your API key within quotations since the expected data type for the API key is a string.
from labelbox import Client
if __name__ == '__main__':
API_KEY = "<your_api_key_here>"
client = Client(API_KEY)
Next steps
Now that you have accessed the Labelbox API via the Python client, it is time to create your first project.
