Accelerated DataFrame Processing

• Speed up the extract, transform, and load (ETL) process for large DataFrames.
• Automatically use all of the processing cores available on your machine.

Optimized for Intel Hardware

• Scale to terabytes of data on a single data science workstation.
• Analyze large datasets (over one billion rows) using heterogeneous data kernels (HDK) and performant end-to-end analytics frameworks that take advantage of the compute power for current and future Intel hardware.

Compatible with Existing APIs and Engines

• Change one line of code to use your existing pandas API calls, no matter the scale. Instead of import pandas as pd use import modin.pandas as pd
• Use Dask*, Ray, or HEAVY.AI* compute engines to distribute the data without having to write code.
• Continue to use the rest of your Python ecosystem code, such as NumPy, XGBoost, and scikit-learn*.
• Use the same notebook to scale from your local machine to the cloud.

Use Case: Fraud Detection

Follow this step-by-step tutorial to learn how to use Intel Distribution of Modin to preprocess, analyze, and transform a credit card transaction dataset for use in a fraud detection application.

Read

Seamlessly Scale pandas Workloads with a Single Code-Line Change

Learn how the Intel Distribution of Modin scales pandas workloads using the same APIs, with a live demonstration that walks you through the tools and process.

Watch

Intel and Anaconda*: Python* Data Science at Scale

Intel and Anaconda have partnered to bring high-performance Python optimizations with simple installations. With minimal code changes, you can accelerate preprocessing, model training, and model inference. See the power of this end-to-end solution in action using the New York City Taxi dataset.

Watch

Scale Your pandas Workflow with Modin

Data scientists no longer have to learn new APIs and rewrite code when their datasets require parallel processing or terabytes of data. See benchmark results that show speedup results for a variety of datasets.

Read

Data Science at Scale with Modin

Get started with the Intel Distribution of Modin by installing via your preferred method, then changing your pandas import statement to use Modin. This brief tutorial then shows how to try it yourself using the New York City Taxi dataset.

Read

Scale Interactive Data Science with Modin and Ray

Learn about the technology that underpins the ability of Modin to scale, how to apply Modin in practice, and how it compares to alternative solutions.

Watch

Unleash The Power Of Dataframes At Any Scale With Modin

This podcast provides insight to how Modin is architected to scale from small to large volumes of data, and how to get started using it.

Listen

The Modin View of Scaling pandas

This article provides background on the vision for Modin, some of the key architectural decisions, and compares Modin to alternative solutions for distributed data processing.

Read

All AI Development Tools and Resources