This repository contains the codes for the paper "Interpreting Deepcode, a learned feedback code". It is a non-linear interpretable model for AWGN channel with feedback.
The noiseless_feedback
folder contains:
deepcode
- Implementation of the Deepcode using Pytorch based on Tensorflow Deepcode.interpretable
- Interpretable model based on Deepcode: 5 hidden states (encoder + decoder) / 7 hidden states (encoder)equivalent
- Contains equivalent expression of interpretable model
The noisy_feedback
folder contains the model when the feedback is noisy.
Python environment:
- Python 3.10.8
- numpy: 1.23.4
- pytorch: 1.12.1
original TensorFlow Deepcode: https://github.com/hyejikim1/Deepcode