Views
No views yet
| Environment | Search | Demo |
|---|---|---|
| TSP | Greedy |  |
| CVRP | Multi-Greedy | |
1conda env create -n <env name> -f environment.yml
2# The environment.yml was generated from
3# conda env export --no-builds > environment.ymlwandb1./rlor
2├── envs
3│ ├── tsp_data.py # load pre-generated data for evaluation
4│ ├── tsp_vector_env.py # define the (vectorized) gym environment
5│ ├── cvrp_data.py
6│ └── cvrp_vector_env.py
7├── models
8│ ├── attention_model_wrapper.py # wrap refactored attention model to cleanRL
9│ └── nets # contains refactored attention model
10└── ppo_or.py # implementaion of ppo with attention model for operation research problems1# apt install diff
2diff --color ppo.py ppo_or.pypython ppo_or.py --num-steps 51 --env-id tsp-v0 --env-entry-point envs.tsp_vector_env:TSPVectorEnv --problem tsppython ppo_or.py --num-steps 60 --env-id cvrp-v0 --env-entry-point envs.cvrp_vector_env:CVRPVectorEnv --problem cvrppython ppo_or.py ... --track--track argument to enable tracking with WandB.