Overview

This is the Pytorch implementation of "Dilated convolution based CSI Feedback Compression for Massive MIMO"

Requirements

To use this project, you need to ensure the following requirements are installed.

The used COST2100 dataset can be found in the paper of CsiNet+ and the corresponding GoogDrive.

python main.py --gpu 0 --lr 2e-3 -v 5 --cr 4 --scenario "in" --expansion 1

or ignore the output

nohup python main.py --gpu 0 --lr 2e-3 -v 5 --cr 4 --scenario "in" --expansion 1  > /dev/null 2>&1 &

The training logs and checkpoints are saved in ./outputs

The checkpoints will be available in GoogleDrive soon.