-
This is the implementation of PCNNs of our paper ("Defining the Benchmarks for a Global Extraction of Information from Deeply Virtual Exclusive Scattering"), https://arxiv.org/abs/2207.10766 .
-
PCNNs is a neural network that incorporates some of the physics knowledge. In particular: error bars and Parity Invariance constraints.
- The code is written in Python=3.8.6, with tensorflow== 2.9.2
- The list of all libraries with the dependencies is in the file:
requirements.txt
- Install the python libraries.
- Download the code from GitHub:
git clone https://github.com/Malmaeen/FemtoNet_PCNNs.git
cd FemtoNet_PCNNs
- To reproduce the results of the DVCS data with a saved model, run the python script:
python driver.py