This repository is not meant to be a source code, rather it tries to publicize the application of data science in a very old field, Power Utilities. Our colleagues and I have performed several years of research and published its results. Since the code parts are confidential, I will try to give you an understanding of the phenomenon and shed light on another possible direction for data scientists to look for oppurtunities.
The provided jupyter notebooks are the explanation of two of our articles: this and this.
Here are the link to each notebook:
1- Problem definition- 7 minutes,
2- Our proposed machine learning pipeline- 7 minutes,
3- Evaluating ours and other techniques, 5 minutes.
Please email me if you have any questions on this topic (my email my last name (ghaderi) dot my firstname (amin) at gmail dot com).