NVIDIA-Merlin/dataloader

The merlin dataloader lets you rapidly load tabular data for training deep leaning models with TensorFlow, PyTorch or JAX

39
/ 100
Emerging

This tool helps machine learning engineers efficiently train recommendation systems. It takes large tabular datasets, often stored in formats like Parquet, and feeds them directly to deep learning models in TensorFlow, PyTorch, or JAX. The result is significantly faster model training, especially for datasets too large to fit into system memory.

423 stars. No commits in the last 6 months.

Use this if you are a machine learning engineer working on recommendation systems and experiencing bottlenecks when loading large tabular datasets for model training.

Not ideal if your primary task doesn't involve training deep learning recommendation models or if your datasets are small enough to be handled efficiently by standard framework dataloaders.

recommender-systems deep-learning-training large-scale-data machine-learning-engineering data-loading-optimization
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

423

Forks

27

Language

Python

License

Apache-2.0

Last pushed

Apr 16, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/NVIDIA-Merlin/dataloader"

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