AIMCEG

Welcome to AIMCEG

AIMCEG stands for Artificial Intelligence based Monte Carlo Event Generator. It is an AI based toolkit to train event generators capable to reproduce physics distributions. We provide a pre-trained generator that you can use to generate new events or you can train it on your own dataset to represent a different distribution.

Requirements

You need python3 in order to run this toolkit. Once you have python3 setup you can easily install all the required python libraries using requirements.txt file included in our code with the following steps.

How to Install AIMCEG?

In order to use this framework, you need to clone the repository and enter the root directory where you will create your scripts/notebooks. Use the following commands

git clone https://github.com/JeffersonLab/AIMCEG.git
cd AIMCEG
pip3 install -r requirements.txt

Note-You might want to use just pip instead of pip3 if you have only python3 installed. In this directory (root AIMCEG directory) you can either open a jupyter notebook or create .py script to import the EventGenerators as shown below.

from AIMCEG.ThreeParticlesEvent import EventGenerator

You can also override our generator for advance utilities (described in tutorials section) as

class MyGenerator(EventGenerator):
    
    def __init__(self, filePath):
        """ filePath is the path to datafile. """
        
        self.mass = mass
        super().__init__(filePath, ignoreColumns=ignoreColumns, computeParticle=computeParticle)

Now you can create an object of your MyGenerator class.

gen = MyGenerator(filePath="eventData.csv")

Once you have this set up, please have a look at instructions for details about supported datafiles.