Skip to content

Attention learn to solve routing problems, TensorFlow2 (TF2), PyTorch, Capacitated Vehicle Routing Problem (CVRP), Transformer, Multi Head Attention, Deep Reinforcement Learning (DRL) (Rollout baseline, REINFORCE) (https://arxiv.org/pdf/1803.08475.pdf)

License

Notifications You must be signed in to change notification settings

daunfamily/VRP_MHA

Repository files navigation

TF2 CVRP solver with Multi Heads Attention

TensorFlow implementation of ATTENTION, LEARN TO SOLVE ROUTING PROBLEMS!(https://arxiv.org/pdf/1803.08475.pdf)

Description

Slide Share -- CVRP solver with Multi Heads Attention --

Dependencies

  • Python >= 3.6
  • TensorFlow = 2.1
  • tqdm
  • scipy
  • numpy
  • plotly (only for plotting)
  • matplotlib (only for plotting)

Usage

First generate the pickle file contaning hyperparameter values by running the following command.

python config.py

then, train the model.

python train.py -p './Pkl/***.pkl'

Plot prediction of the pretrained model

python plot.py -p './Weights/***.h5'

About

Attention learn to solve routing problems, TensorFlow2 (TF2), PyTorch, Capacitated Vehicle Routing Problem (CVRP), Transformer, Multi Head Attention, Deep Reinforcement Learning (DRL) (Rollout baseline, REINFORCE) (https://arxiv.org/pdf/1803.08475.pdf)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%