Release of AutoGMap:Learning to Map Large-scale Sparse Graphs on Memristive Crossbar

For reproduction of our searched model, the Ubuntu OS is recommended. The models have been tested using Python 3.6 + Pytorch 1.0.0

Required packages:

python=3.6.2
scipy=1.5.1
tensorboard=1.14.0
tensorboardx=2.2
torch=1.0.0
numpy=1.17.4

Visualization of block Coverage during training on qh882:

Visualization of block Coverage during training on qh1484