Dask

Samples of code for common Dask user tasks

Learn distributed computing with Dask to scale your Python analysis across multiple machines and process datasets that don’t fit in memory.

Getting started with Dask

New to Dask? Start with Getting started with Dask to learn the basics of creating and using Dask clusters.

Working with data? Combine Dask with your data loading workflows to process larger datasets that don’t fit on a single machine.

Training models? Use Dask with PyTorch to distribute model training across multiple GPUs and machines.

When to use Dask

Dask is perfect when your current Python code becomes too slow or runs out of memory. It works with pandas, NumPy, and scikit-learn to scale your existing workflows with minimal code changes.