https://github.com/ansebi/sin-activated_neural_network_for_time_series_forecasting

What if use sin activation for predicting waves? Here you can find: 1) wave data generator 2) sin-based nn 3) tests both on a sin wave and on the real data
https://github.com/ansebi/sin-activated_neural_network_for_time_series_forecasting

Last synced: about 2 months ago
JSON representation

What if use sin activation for predicting waves? Here you can find: 1) wave data generator 2) sin-based nn 3) tests both on a sin wave and on the real data

• Host: GitHub
• URL: https://github.com/ansebi/sin-activated_neural_network_for_time_series_forecasting
• Owner: Ansebi
• Created: 2023-08-03T08:15:16.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-08-26T07:19:15.000Z (9 months ago)
• Last Synced: 2025-01-21T12:35:54.486Z (4 months ago)
• Language: Jupyter Notebook
• Size: 4.55 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# SIN-activated Neural Network for Time Series Forecasting

What if use sin activation for predicting waves?


Here you can find:


1) wave data generator


2) sin-based nn


3) tests both on a sin wave and on the real data



The multiwave case in ipynb is the way to go: the most universal and concise.


The rest of the notebooks may be considered as an illustration.


Multiwave notebook requires sin_nn module (the model itself)


and the metrics module.