https://github.com/rodrigobressan/entity_embeddings_categorical
Discover relevant information about categorical data with entity embeddings using Neural Networks (powered by Keras)
https://github.com/rodrigobressan/entity_embeddings_categorical
categorical-data embeddings entity-embedding keras machine-learning neural-networks pre-processing utility-library
Last synced: about 1 month ago
JSON representation
Discover relevant information about categorical data with entity embeddings using Neural Networks (powered by Keras)
- Host: GitHub
- URL: https://github.com/rodrigobressan/entity_embeddings_categorical
- Owner: rodrigobressan
- License: mit
- Created: 2019-02-22T21:03:27.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2022-12-08T01:38:04.000Z (over 2 years ago)
- Last Synced: 2025-03-27T01:35:54.601Z (2 months ago)
- Topics: categorical-data, embeddings, entity-embedding, keras, machine-learning, neural-networks, pre-processing, utility-library
- Language: Python
- Homepage:
- Size: 6.56 MB
- Stars: 70
- Watchers: 3
- Forks: 19
- Open Issues: 16
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
[](https://pypi.org/project/entity-embeddings-categorical)
[](https://travis-ci.org/bresan/entity_embeddings_categorical)
[](https://coveralls.io/github/bresan/entity_embeddings_categorical?branch=master)
[](https://github.com/bresan/entity_embeddings_categorical/blob/master/LICENSE.md)
[](https://www.codacy.com/app/bresan/entity_embeddings_categorical?utm_source=github.com&utm_medium=referral&utm_content=bresan/entity_embeddings_categorical&utm_campaign=Badge_Grade)
# Overview
This project is aimed to serve as an utility tool for the preprocessing, training and extraction of entity embeddings through Neural Networks using the Keras framework. It's still under construction, so please use it carefully.
# Installation
The installation is pretty simple if you have a virtualenv already installed on your machine. If you don't please rely to [VirtualEnv official documentation](https://virtualenv.pypa.io/en/latest/).
```bash
pip install entity-embeddings-categorical
```
# Documentation
Besides the docstrings, major details about the documentation can be found [here](https://entity-embeddings-categorical.readthedocs.io/en/latest/).
# Testing
This project is inteded to suit most of the existent needs, so for this reason, testability is a major concern. Most of the code is heavily tested, along with [Travis](https://travis-ci.org/bresan/entity_embeddings_categorical) as Continuous Integration tool to run all the unit tests once there is a new commit.
# Usage
The usage of this utility library is provided in two modes: default and custom. In the default configuration, you can perform the following operations: Regression, Binary Classification and Multiclass Classification.
If your data type differs from any of these, you can feel free to use the custom mode, where you can define most of the configurations related to the target processing and output from the neural network.
## Default mode
The usage of the default mode is pretty straightforward, you just need to provide a few parameters to the Config object:
So for creating a simple embedding network that reads from file **sales_last_semester.csv**, where the target name is **total_sales**, with the desired output being a **binary classification** and with a training ratio of **0.9**, our Python script would look like this:
```python
config = Config.make_default_config(csv_path='sales_last_semester.csv',
target_name='total_sales',
target_type=TargetType.BINARY_CLASSIFICATION,
train_ratio=0.9)
embedder = Embedder(config)
embedder.perform_embedding()
```
Pretty simple, huh?
A working example of default mode can be found [here as a Python script](https://github.com/bresan/entity_embeddings_categorical/blob/master/example/default/default_config_example.py).
## Custom mode
If you intend to customize the output of the Neural Network or even the way that the target variables are processed, you need to specify these when creating the configuration object.
This can be done by creating a class that extend from [TargetProcessor](https://github.com/bresan/entity_embeddings_categorical/blob/master/entity_embeddings/processor/processor.py) and [ModelAssembler](https://github.com/bresan/entity_embeddings_categorical/blob/master/entity_embeddings/network/assembler.py).
A working example of custom configuration mode can be found [here](https://github.com/bresan/entity_embeddings_categorical/blob/master/example/custom/custom_config_example.py).
## Visualization
Once you are done with the training of your model, you can use the module [visualization_utils](https://github.com/bresan/entity_embeddings_categorical/blob/master/entity_embeddings/util/visualization_utils.py) in order to create some visualizations from the generated weights as well as the accuraccy of your model.
Below are some examples created for the [Rossmann dataset](https://www.kaggle.com/c/rossmann-store-sales):
# Troubleshooting
In case of any issue with the project, or for further questions, do not hesitate to open an issue here on GitHub.
# Contributions
Contributions are really welcome, so feel free to open a pull request :-)
# TODO
- Allow to use a Pandas DataFrame instead of the csv file path;