hipDF documentation#
2025-05-05
1 min read
Note
hipDF is in an early access state. Running production workloads is not recommended.
Based on the Apache Arrow columnar memory format, the hipDF library enables DataFrames that let you load and manipulate your data in the memory of AMD GPUs. Following a familiar Pandas-like API, hipDF lets you accelerate both your existing and new data science workloads on AMD GPUs with high-level functions that eliminate the need to go into low-level HIP programming.
In addition to containing all of the necessary tools to build powerful new data processing applications, hipDF enables you to easily port your existing Pandas and cuDF workloads to AMD GPUs. hipDF is derived from the NVIDIA RAPIDS® open-source project cuDF.
The hipDF code is open and hosted at ROCm-DS/hipDF.
The hipDF documentation is structured as follows:
To contribute to the documentation refer to Contributing to ROCm-DS.
You can find licensing information on the Licensing page.