From 44a544ebb7818ae5b230e6e91014afae7e948c9c Mon Sep 17 00:00:00 2001 From: Sanjana S <158697809+SanjanaS10@users.noreply.github.com> Date: Sun, 11 Jan 2026 22:07:20 +0530 Subject: [PATCH] Improve README formatting and feature descriptions Updated README for clarity and formatting improvements. --- .../README.md | 36 ++++++++++++------- 1 file changed, 23 insertions(+), 13 deletions(-) diff --git a/Graph_Representation_Learning_Rushil_Singha/README.md b/Graph_Representation_Learning_Rushil_Singha/README.md index ead8c90..653a8e9 100644 --- a/Graph_Representation_Learning_Rushil_Singha/README.md +++ b/Graph_Representation_Learning_Rushil_Singha/README.md @@ -7,27 +7,29 @@ This project builds **k-nearest neighbor (kNN) jet graphs**, learns **Chebyshev --- ## 🚀 Features + - kNN graph construction from jet particle clouds - Graph encoder using **Chebyshev GCN** (`SimpleChebNet`) - Latent **diffusion process** with denoising MLP - Jet particle **decoder** network -- Evaluation with **KL divergence** & **Wasserstein distance** +- Evaluation using **KL divergence** & **Wasserstein distance** - Visualization utilities for jet properties --- ## ⚙️ Installation -Clone the repo and install dependencies: +Clone the repository and install dependencies: -```bash + +```sh git clone https://github.com/your-username/jetnet-graph-diffusion.git cd jetnet-graph-diffusion - pip install -r requirements.txt +``` requirements.txt - +```sh numpy==1.24.3 torch==2.0.0 torch-geometric @@ -38,18 +40,26 @@ networkx scikit-learn jetnet ``` -# This script: -->Encodes jets into latent space +## 📜 Script Overview -->Runs diffusion training +The main script performs the following steps: -->Decodes jets back into particle space - -->Logs evaluation metrics - -->Saves visualizations to results/ +- Encodes jets into latent space +- Runs diffusion training +- Decodes jets back into particle space +- Logs evaluation metrics +- Saves visualizations to results/ directory +## To run the main workflow: +```sh +python code.py +``` +## Usage +This repository contains a research prototype for graph-based jet generation using latent diffusion models. +## Note: +Dataset paths, hyperparameters, and output locations may need to be adjusted depending on the local environment and JetNet configuration. +