A benchmark to the challenge of compressing an obscured dataset containing Particle Identification
The data can be found here: https://zenodo.org/record/1231531#.WyZSQFOFO3V
Download it and put it into a folder called 'Data' in this directory.
Installing FunctionScaler(https://github.com/weissercn/FunctionScaler/) as a dependency is required. This can be done using pip (pip install FunctionScaler). Other requirements: keras, sklearn, pandas, matplotlib, pickle
Run the following notebook in this order:
- Prepare
- Train
- Analyse Output
- Cross Check
- ROOT compression