Overview This repository contains code and resources for time series forecasting using various machine learning techniques. The goal is to predict future values based on historical data, leveraging models such as LSTM neural networks, ARIMA, and ensemble methods.
Features Data preprocessing and feature engineering Implementation of multiple forecasting models LSTM neural networks ARIMA XGBoost and other ensemble methods Model evaluation and comparison Visualization of results Datasets One of the main datasets used in this project is customer billing data. To ensure privacy and reproducibility, the repository includes a script to create a pseudo dataset that mimics the characteristics of the original customer billing data.
Installation To get started, clone the repository and install the required dependencies:
- git clone https://github.com/RonKlem/time_series_forecasting.git
- cd time_series_forecasting
- pip install -r requirements.txt
Usage
Data Wrangling: Run the data wrangling script to clean and prepare the data.
- python customer_billing_data_wrangling.ipynb
Model Training: Train the forecasting models using the prepared data.
- python Forecasting_POC.ipynb
(not available yet) Evaluation: Evaluate the performance of the models and visualize the results.
- python evaluate.py
(not available yet) Results The results of the forecasting models, including performance metrics and visualizations, will be saved in the results/ directory.
Contributing Contributions are welcome! Please open an issue or submit a pull request if you have any suggestions or improvements.
Contact For any questions or inquiries, please contact Ron Klem at [email protected]