Optimization Methods for Track Fitting in the Active-Target Time Projection Chamber

This repository contains the code made in 2018-2019 for analyzing global and local optimization methods for the track fitting process.

Packages

Following Python packages are required for using the code:

• pytpc
• numpy
• SciPy
• matplotlib
• Keras
• tensorflow

The code is written in Python 3.6.

Global Track-Fitting Methods

• naive Monte Carlo method
• differential evolution
• basin hopping

Folders

• hpc-scripts

  • Contains shell scripts for hpc job submission
  • Contains Python files for MC fitting

• jupyter notebooks

  • Contains the code used to analyze different real and simulated proton events
  • Contains the plots of track fitting using different global and local optimization methods
  • Contains the Monte-Carlo (MC) Algorithm written entirely in Python, and MC fitting of the original and modified objective functions

• proton-classification

  • Contains codes for using Keras model to classify proton events