DeepMuon 1.9.1

DeepMuon Project Built for Simple and Direct Deep Learning Researches on Dark Matter Searching

Introduction

DeepMuon is a easy-using deep learning platform built for dark matter searching experiments, we are eager to provide advanced model training frame and excellent project management assistance.

Here we list out the available features of DeepMuon:

• DeepMuon support Single GPU training and Distributed Data Parallel training
• Model Inference and analysis
• Customize models
• Customize datasets
• Customize loss functions
• Easy to train/inference
• Tidy file system
• Simple and direct tutorials

Installation

git clone https://github.com/Airscker/DeepMuon.git
cd DeepMuon
pip install -v -e ./ --user

Tutorials

• Regulations
• [Customize Dataset](./Tutorial/2.Customize Dataset.md)
• [Customize Model](./Tutorial/3.Customize Model.md)
• Configuration
• [Training and Inference](./Tutorial/5.Training and Inference.md)

CopyRight

GNU AFFERO GENERAL PUBLIC LICENSE

Project: DeepMuon

Dark Matter Searching Deep Learning Platform

Author: Airscker/Yufeng Wang

Contributors: Yufeng Wang, Yu Zhang

University of Science of Technology and Science of China

Internal supported sensor's dataset: Pandax-4T, Hailing

If you want to publish thesis using DeepMuon, please add bibliography:

To be published

Copyright (C) 2022 by Airscker, All Rights Reserved.