This script processes pickle files from the ColumnFlow workflow (specifically from the mergeHistograms step). It takes histogram data from different reconstruction levels and combines them into a single plot for comparison.
- Reads multiple pickle files containing histograms from different reconstruction levels.
- Handles different binning structures.
- Normalizes all histograms so that their event counts are scaled between 0 and 1 on the y-axis.
- Generates a combined plot for visual comparison.
- Ensure you have the required pickle files from the ColumnFlow workflow.
- Run the script with the paths to the relevant files as input.
- View the output plot to compare the histograms.