DataRobot supports REST, Python, and R APIs as a programmatic alternative to the UI for creating and managing DataRobot projects. It allows you to automate processes and iterate more quickly, and lets you use DataRobot with scripted control. The API provides an intuitive modeling and prediction interface. You can use the API with DataRobot-supported clients in either R or Python, or with your own custom code. The clients are supported in Windows, UNIX, and OS X environments. Additionally, you can generate predictions with the prediction and batch prediction APIs, and build DataRobot blueprints in the blueprint workshop.
Review the sections of the API documentation below to find the best resources for your needs:
New users can review the API quickstart guide to configure their environments and get started with DataRobot's APIs.
The API user guide details examples of machine learning workflows and use cases that address common data science problems.
The API reference provides documentation on the DataRobot REST API, Python client, R client, blueprint workshop, Prediction API, and Batch Prediction API.
Browse user guide topics to find complete examples of common data science and machine learning workflows. The API user guide includes overviews, Jupyter notebooks, and task-based tutorials.
The DataRobot REST API provides a programmatic alternative to the UI for creating and managing DataRobot projects. Note that accessing the legacy REST API docs requires you to be logged into the DataRobot application.
DataRobot customers can reference the OpenAPI specification for the DataRobot REST API. It helps automate the generation of a client for languages that DataRobot doesn't offer a client. It assists with design, implementation, and testing integration with DataRobot's REST API using a variety of automated OpenAPI-compatible tools. It describes, if any, feature flags that are necessary for use of a given endpoint.
AI Accelerators are designed to help speed up model experimentation, development, and production using the DataRobot API. They codify and package data science expertise in building and delivering successful machine learning projects into repeatable, code-first workflows and modular building blocks. AI Accelerators are ready right out-of-the-box, work with the notebook of your choice, and can be combined to suit your needs.
The table below outlines which versions of DataRobot's SDKs correspond to DataRobot's Self-Managed AI Platform versions.
Self-Managed AI Platform version
Python SDK version
R SDK version
v10.2
v3.6
v2.18.6 (GA)
v2.31.2 (preview)
v10.1
v3.5
v2.18.6 (GA)
v2.31.2 (preview)
v10.0
v3.4
v2.18.6 (GA)
v2.31.2 (preview)
v9.2
v3.3
v2.18.6 (GA)
v2.31.2 (preview)
v9.1
v3.2
v2.18.4 (GA)
v2.31.2 (preview)
v9.0
v3.1
v2.29 (preview)
v8.0
v2.28
v2.18.2
v7.3
v2.27.3
v2.18.2
v7.2
v2.26.0
v2.18.2
v7.1
v2.25.1
v2.18.2
v7.0
v2.24.0
v2.18.2
v6.3
v2.23.0
v2.17.1
v6.2
v2.22.1
v2.17.1
v6.1
v2.21.5
v2.17.1
v6.0
v2.20.2
v2.17.1
v5.3
v2.19.0
v2.17.1
v5.2
v2.18.0
v2.17.1
v5.1
v2.17.0
v2.17.1
v5.0
v2.15.1
v2.15.0
v4.5
v2.14.2
v2.14.2
v4.4
v2.13.3
v2.13.1
v4.3
v2.11.2
v2.11.0
v4.2
v2.9.3
v2.9.0
v4.0
v2.8.3
v2.8.0
v3.1
v2.7.3
v2.7.1
v3.0
v2.6.2
v2.6.0
v2.9
v2.4.3
v2.4.0
v2.8
v2.0.37
v2.0.30
Note
Both the backend and clients use versioning in the format Major.Minor.Patch (e.g., v2.3.1), but there is no relationship between the patch version of the backend and the patch version of the clients. There is a requirement, however, that the backend version has a major.minor version equal to or greater than the client version. For example, a v2.2 client can "talk" to either a v2.2 backend or a v2.4 backend, but cannot be used with a v2.0 backend.
mkdir -p ~/datarobot_2.18.2 && tar -xvzf ~/Downloads/datarobot_2.18.2.tar.gz -C ~/datarobot_2.18.2
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.18.2/datarobot')
mkdir -p ~/datarobot_2.18.2 && tar -xvzf ~/Downloads/datarobot_2.18.2.tar.gz -C ~/datarobot_2.18.2
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.18.2/datarobot')
Python: pip install "datarobot>=2.26.0,<2.27"
R:
mkdir -p ~/datarobot_2.18.2 && tar -xvzf ~/Downloads/datarobot_2.18.2.tar.gz -C ~/datarobot_2.18.2
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.18.2/datarobot')
Python: pip install "datarobot>=2.25.1,<2.26"
R:
mkdir -p ~/datarobot_2.18.2 && tar -xvzf ~/Downloads/datarobot_2.18.2.tar.gz -C ~/datarobot_2.18.2
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.18.2/datarobot')
Python: pip install "datarobot>=2.24.0,<2.25.1"
R:
mkdir -p ~/datarobot_2.18.2 && tar -xvzf ~/Downloads/datarobot_2.18.2.tar.gz -C ~/datarobot_2.18.2
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.18.2/datarobot')
mkdir -p ~/datarobot_2.17.1 && tar -xvzf ~/Downloads/datarobot_2.17.1.tar.gz -C ~/datarobot_2.17.1
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.17.1/datarobot')
Python: pip install "datarobot>=2.22.1,<2.23"
R:
mkdir -p ~/datarobot_2.17.1 && tar -xvzf ~/Downloads/datarobot_2.17.1.tar.gz -C ~/datarobot_2.17.1
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.17.1/datarobot')
Python: pip install "datarobot>=2.21.5,<2.22.1"
R:
mkdir -p ~/datarobot_2.17.1 && tar -xvzf ~/Downloads/datarobot_2.17.1.tar.gz -C ~/datarobot_2.17.1
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.17.1/datarobot')
Python: pip install "datarobot>=2.20.2,<2.21.5"
R:
mkdir -p ~/datarobot_2.17.1 && tar -xvzf ~/Downloads/datarobot_2.17.1.tar.gz -C ~/datarobot_2.17.1
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.17.1/datarobot')
mkdir -p ~/datarobot_2.17.1 && tar -xvzf ~/Downloads/datarobot_2.17.1.tar.gz -C ~/datarobot_2.17.1
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.17.1/datarobot')
Python: pip install "datarobot>=2.18,<2.19"
R:
mkdir -p ~/datarobot_2.17.1 && tar -xvzf ~/Downloads/datarobot_2.17.1.tar.gz -C ~/datarobot_2.17.1
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.17.1/datarobot')
Python: pip install "datarobot>=2.17,<2.18"
R:
mkdir -p ~/datarobot_2.17.1 && tar -xvzf ~/Downloads/datarobot_2.17.1.tar.gz -C ~/datarobot_2.17.1
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.17.1/datarobot')
Python: pip install "datarobot>=2.15,<2.16"
R:
mkdir -p ~/datarobot_2.15.0 && tar -xvzf ~/Downloads/datarobot_2.15.0.tar.gz -C ~/datarobot_2.15.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.15.0/datarobot')
mkdir -p ~/datarobot_2.14.0 && tar -xvzf ~/Downloads/datarobot_2.14.0.tar.gz -C ~/datarobot_2.14.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.14.0/datarobot')
Python: pip install "datarobot>=2.13,<2.14"
R:
mkdir -p ~/datarobot_2.13.0 && tar -xvzf ~/Downloads/datarobot_2.13.0.tar.gz -C ~/datarobot_2.13.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.13.0/datarobot')
Python: pip install "datarobot>=2.12,<2.13"
R:
mkdir -p ~/datarobot_2.12.1 && tar -xvzf ~/Downloads/datarobot_2.12.1.tar.gz -C ~/datarobot_2.12.1
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.12.1/datarobot')
Python: pip install "datarobot>=2.11,<2.12"
R:
mkdir -p ~/datarobot_2.11.0 && tar -xvzf ~/Downloads/datarobot_2.11.0.tar.gz -C ~/datarobot_2.11.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.11.0/datarobot')
Python: pip install "datarobot>=2.9,<2.10"
R:
mkdir -p ~/datarobot_2.9.0 && tar -xvzf ~/Downloads/datarobot_2.9.0.tar.gz -C ~/datarobot_2.9.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.9.0/datarobot')
Python: pip install "datarobot>=2.8,<2.9"
R:
mkdir -p ~/datarobot_2.8.0 && tar -xvzf ~/Downloads/datarobot_2.8.0.tar.gz -C ~/datarobot_2.8.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.8.0/datarobot')
mkdir -p ~/datarobot_2.7.0 && tar -xvzf ~/Downloads/datarobot_2.7.0.tar.gz -C ~/datarobot_2.7.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.7.0/datarobot')
Python: pip install "datarobot>=2.8,<2.9"
R:
mkdir -p ~/datarobot_2.6.0 && tar -xvzf ~/Downloads/datarobot_2.6.0.tar.gz -C ~/datarobot_2.6.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.6.0/datarobot')
mkdir -p ~/datarobot_2.4.0 && tar -xvzf ~/Downloads/datarobot_2.4.0.tar.gz -C ~/datarobot_2.4.0
install.packages('devtools') # (If you don't already have devtools on your system.)
devtools::install('~/datarobot_2.4.0/datarobot')