Skip to content

Implementation of Unified Embedding: Battle-Tested Feature Representations for Web-Scale ML Systems

License

Notifications You must be signed in to change notification settings

ledmaster/unified-embeddings

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Unified Embeddings PyTorch Implementation

This project is a simple implementation of a Unified Embedding paper.

The model uses a single embedding table to encode multiple different categorical features in order to save memory and computation time when dealing with large-scale systems.

You can find a detailed explanation of the code here.

Installing / Getting started

To get started with this project, you need to have Python and PyTorch installed. You can install the required packages using pip:

pip install torch polars xxhash

The above command installs PyTorch, Polars, and xxhash.

PyTorch is used for creating and training the neural network model, Polars for data manipulation, and xxhash for hashing.

Features

The main features of this project include:

  • Unified embedding layer (ue.py)
  • Simple feed-forward neural network for prediction (test.py)
  • Training and validation loop for testing the code (test.py)

Links

Licensing

The code in this project is licensed under MIT license.

About

Implementation of Unified Embedding: Battle-Tested Feature Representations for Web-Scale ML Systems

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages