kklemon/FlashPerceiver

Fast and memory efficient PyTorch implementation of the Perceiver with FlashAttention.

24
/ 100
Experimental

For machine learning engineers working with large and complex datasets, FlashPerceiver provides a highly efficient way to process data using Perceiver models. It takes your raw data, potentially with various input dimensions and masks, and processes it into compressed, meaningful representations or task-specific outputs, even for multi-task scenarios. This project is ideal for those who need to build or train deep learning models that can handle very long input sequences without prohibitive memory or speed costs.

No commits in the last 6 months.

Use this if you are a machine learning engineer or researcher building Perceiver-based models and need significantly faster training times and reduced memory consumption when dealing with high-dimensional or long sequence data.

Not ideal if you are looking for a pre-trained model or a high-level API for non-machine learning tasks, as this is a foundational library for model implementation.

deep-learning large-scale-data model-training sequence-processing computational-efficiency
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 8 / 25
Community 9 / 25

How are scores calculated?

Stars

32

Forks

3

Language

Python

License

Last pushed

Nov 04, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/kklemon/FlashPerceiver"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.