/demo-arize

Primary LanguageJupyter Notebook

This repository contains a notebook ArizeDemo.ipynb, which demonstrates one approach for integrating Arize with Algorithmia. The workflow contains the following steps:

  • Train an sklearn model.
  • Upload the model to a hosted data collection on Algorithmia.
  • Call the model for inference by piping model input into the sklearn_demo algorithm, which is hosted on Algorithmia's public cluster. The algorithm is set up to log inference metrics to Arize before returning a response.