Deep Learning for Real-Time Top Quark Jet Tagging

End-to-end MATLAB Deep Learning workflow for Real-Time Top Quark Jet Tagging is presented
37 Downloads
Updated 1 Feb 2023

Deep-Learning-for-Real-Time-Top-Jet-Tagging

End-to-end MATLAB® workflow for Real-Time Top Quark Jet Tagging is presented. Live script contains a predictive model, based on deep convolutional neural network, that discriminates top quark (signal) jets from QCD plain vanilla (background) jets. Besides a predictive model, the workflow presented includes: accessing and preprocessing particle scattering data, transforming jets to 2D images, and code generation for deployment of the network on FPGA.

Setup

To Run:

  1. Download particle jets open datasets as instructed in the Reference Datasets section of the Live script. Open Python, import part of the randomly sampled data as pandas dataframes and save in parquet format.
  2. Import parquet data as a MATLAB table, preprocess jets to images and save to disc.
  3. Build deep convolusional neural network using App designer® and train network using training datasets.
  4. Check accuracy of the network on test datasets.
  5. Deploy trained network on FPGA following Deploy Trained Network on FPGA section of the Live script.

MathWorks Products (https://www.mathworks.com)

Requires MATLAB release R2020a or newer

Cite As

Temo Vekua (2024). Deep Learning for Real-Time Top Quark Jet Tagging (https://github.com/MathWorks-Teaching-Resources/Deep-Learning-for-Real-Time-Top-Jet-Tagging), GitHub. Retrieved .

MATLAB Release Compatibility
Created with R2021b
Compatible with any release
Platform Compatibility
Windows macOS Linux
Tags Add Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Versions that use the GitHub default branch cannot be downloaded

Version Published Release Notes
1.2.0

included image

1.1.0

connected to github

1.0.0

To view or report issues in this GitHub add-on, visit the GitHub Repository.
To view or report issues in this GitHub add-on, visit the GitHub Repository.