---

Author: Philipp Ganster

Link to Project: https://gasspeicher-deutschland.herokuapp.com/

---

About this Project

Due to a spike in german gas prices, the aim of this project is
to raise awareness in regard to gas consumption of Germany, by 
showing an overview of all german gas providers and their:

- Storage in TWh
- Capacity Full in %
- Daily Change in Demand in %
- Daily Change in Demain in GWh

Advanced Python Schedular is used for running script 
every day at 8pm, since API Data is refreshed every day at 7:30pm (MESZ)
https://apscheduler.readthedocs.io/en/3.x/

Further information

• Streamlit Documentation: https://docs.streamlit.io/library/api-reference
• API Data: https://agsi.gie.eu/

About the Development Process (Pipeline for this Project)

1. Develop App locally with Pycharm CE
2. Use Streamlit (streamlit run main.py) for Agile Testing
3. Commit and Push Project to GitHub Repository
4. Deploy Repository on GitHub to Heroku
5. Use GitHub Desktop for continuous Development

Information about the Model for Price Prediction

Polynomial Function MAE = 0.008515221391126407
Polynomial Function R Squared = 0.6370797480659653

Function

$f(x_1 , x_2) = 5.02 - (6.69 {e-09}) * x_1 + (1.25 {e-09}) * x_2 + (2.24 {e-18}) * x_1^2 - (7.83 {e-19}) * x_2^2 - (3.19 {e-19}) *x_1 * x_2$

Further Information see Repository: https://github.com/pgphi/GasPred