Get Started

To get started with RIME, first make sure your deployment is installed. Then perform the setup steps.

Set up RIME

Once you have access to a working RIME deployment, you need the following items in order to start performing data science tasks:

  • Get your RIME cluster URL

  • Create an API token

  • Install the RIME SDK

Cluster URL

The URL of your RIME cluster is where the webUI for RIME is available. This URL is also used to upload results. The URL is typically in the form rime.{}.rbst.io. Consult your site administrator with questions about your cluster URL.

API token

You now need to create an API token. This will be used to connect programmatically to the RIME cluster. Creating this API Token is done in the web UI. Go to Settings -> API Access Tokens and create your own token. Full details on API tokens are discussed in Users, authorization, and authentication.

Install the RIME SDK

The RIME SDK is a Python package for interacting with RIME programmatically. This SDK is available on pypi, so it can easily be installed with pip install rime-sdk. This will install the most recent version of the RIME SDK. If your cluster is on an older version of RIME, you can install that specific version with pip install rime-sdk==X.Y.Z.

Set up RIME

Once you have access to a working RIME deployment, you need the following items in order to start performing data science tasks:

  • Get your RIME cluster URL

  • Create an API token in the UI

  • Install the RIME SDK

Cluster URL

The first order of business is figuring out the cluster URL for your RIME cluster. This is the URL at which you will view your results, and is also used to upload your results. It is usually in the form rime.{}.rbst.io. If you have any questions about what your cluster URL is, please ask your administrator (or whoever set up the RIME cluster).

API token

You now need to create an API token. This will be used to connect programmatically to the RIME cluster. Creating this API Token is done in the web UI. Go to Settings -> API Access Tokens and create your own token. Full details on API tokens are discussed in API Authentication.

Install the RIME SDK

The RIME SDK is a Python package for interacting with RIME programmatically. This SDK is available on pypi, so it can easily be installed with pip install rime-sdk. This will install the most recent version of the RIME SDK. If your cluster is on an older version of RIME, you can install that specific version with pip install rime-sdk==X.Y.Z.

Try a tutorial notebook

Each notebook illustrates how to use RIME to validate, protect, and monitor your ML pipeline across a wide range of tasks, from classification to ranking, and across tabular, natural language processing (NLP), and computer vision (CV) data modalities.

The datasets and models used in these examples come from well-known public sources such as

UCI ML or arXiv.

In these notebooks, we walk through the steps of setting up RIME with an example dataset/model; we run Stress Testing first, then Continuous Testing.

Tabular Notebooks

NLP Notebooks

CV Notebooks

Set up a new test run

After the tutorials, you can get familiar with details about starting a new Continuous or Stress test.