The goal of this repository is to provide a data science framework to explore various anomaly detection strategies in the context of the CMS Tracker DQM Playground effort. The code for the website hosting the data and allowing to inspect them can be found here. An optional Command Line Interface to get the data can be found here.
An overview of the architecture of the project highlighting the role of the data science part can be found below.
The current structure can be divided in three steps:
- data processing (dummy)
- data science
- data visualization (dummy)
Another pipeline could be added to interface directly the website's API with the data science framework.