marcusGH/edain_paper

Contains the implementation of the EDAIN and EDAIN-KL methods proposed in our paper. The research was also part of the thesis I wrote as part of my MSc in Statistics (Data Science) at Imperial College London

14
/ 100
Experimental

This project offers a novel way to prepare complex time series data for machine learning models, especially deep neural networks. It takes raw, irregular time series data—like financial trading logs or customer credit histories—and adaptively normalizes it. The output is preprocessed data that significantly improves the accuracy and efficiency of predictions or classifications. This is ideal for data scientists or quantitative analysts who build predictive models from real-world, noisy time series.

No commits in the last 6 months.

Use this if you are building deep learning models for time series data with irregularities like multiple modes, skewness, or outliers, and you need a more effective preprocessing method than standard normalization.

Not ideal if your time series data is already well-behaved, or if you are not using deep neural networks for your prediction or classification tasks.

time-series-forecasting credit-risk-modeling financial-trading data-preprocessing machine-learning-engineering
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 0 / 25

How are scores calculated?

Stars

16

Forks

Language

Jupyter Notebook

License

Last pushed

Feb 19, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/marcusGH/edain_paper"

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