cake-lab/DELI

Optimizing loading training data from cloud bucket storage for cloud-based distributed deep learning. Official repository for Quantifying and Improving Performance of Distributed Deep Learning with Cloud Storage, to be published in IC2E 2021

28
/ 100
Experimental

When training large-scale deep learning models in the cloud, you often store your training data in cloud storage buckets. This project helps deep learning engineers significantly speed up the process of loading this data, reducing the time your training loop spends waiting. It takes your cloud-based training data and outputs faster, more cost-effective distributed deep learning model training. Deep learning engineers and MLOps specialists using cloud infrastructure would benefit from this.

No commits in the last 6 months.

Use this if you are a deep learning engineer training models with large datasets stored in cloud storage buckets, and you're experiencing slow data loading times during distributed training.

Not ideal if your deep learning models are small, you train on a single machine, or your data is primarily stored on local disks rather than cloud buckets.

deep-learning-engineering cloud-ml-training mlops distributed-training data-loading-optimization
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 7 / 25

How are scores calculated?

Stars

11

Forks

1

Language

Jupyter Notebook

License

MIT

Last pushed

Jan 01, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/cake-lab/DELI"

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