zimmerrol/keras-utility-layer-collection

Collection of custom layers and utility functions for Keras which are missing in the main framework.

42
/ 100
Emerging

This collection of Keras layers helps deep learning researchers and practitioners re-implement advanced neural network architectures, especially those involving attention mechanisms. It takes existing Keras models and adds specialized layers like Multi-Head Attention or Layer Normalization to reproduce cutting-edge research. The primary user is a machine learning researcher or engineer working with deep learning models in Keras.

No commits in the last 6 months.

Use this if you are a deep learning researcher or practitioner using Keras and need to implement or replicate advanced 'attention' mechanisms from state-of-the-art research papers.

Not ideal if you are not working with Keras or if your deep learning tasks do not require specialized attention or normalization layers beyond what Keras offers out-of-the-box.

deep-learning-research neural-networks attention-mechanisms AI-model-development machine-learning-engineering
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 18 / 25

How are scores calculated?

Stars

62

Forks

15

Language

Python

License

MIT

Last pushed

May 25, 2020

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/zimmerrol/keras-utility-layer-collection"

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