Below are some example notebooks and accompanying files that demonstrate how you can use the Virtualitics Python API in your workflow. Each file is linked below within its respective project or all files can be found at the bottom of this article.

Exploring eCommerce Travel Data

This example gives an introduction to using the Virtualitics Python API to make functional calls to Virtualitics Explore from your notebook. We focus on loading data, running our built-in AI routines, and leveraging Python to create custom features that can then be fed directly into Virtualitics Explore.

Downloads:

Exploring eCommerce Travel Data - HTML

Exploring eCommerce Travel Data - Jupyter Notebook

Model Visualization and Explainability

In this example, we investigate a more advanced use case of the API by building models in the notebook then visualizing the outputs in Virtualitics Explore. We combine typical data science libraries like scikit-learn and TensorFlow with the advanced visualizations created in Virtualitics Explore to look inside the AI “black box” and understand what the models are learning from the data. Furthermore, we use visualizations to very quickly and easily optimize the model.

Downloads:

Explainable ML - HTML

Explainable ML - Jupyter Notebook

Using Virtualitics Explore's Network Extractor on ETF Data

This notebook demonstrates how you can use Virtualitics Explore's Network Extractor to create a network dataset from a tabular spreadsheet. This can be very useful for analyzing data with many categorical features, and can provide insight about similarities and differences that exist in the data.

Downloads:

Network Extractor with Exchange Traded Funds - HTML

Network Extractor with Exchange Traded Funds - Jupyter Notebook