Skip to content

CIAM-Group/CaDA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CaDA

The official code repository for CaDA: Cross-Problem Routing Solver with Constraint-Aware Dual-Attention.

Preparation

  1. Clone the repository:
git clone https://github.com/CIAM-Group/CaDA.git
  1. Download datasets:
  • Download data.zip from Hugging Face.
  • Unzip data.zip and organize the files in the project directory as follows:
CaDA
├── data
│   ├── lib_data
│   └── synthetic_data
├── 50
├── 100
└── utils
  1. Download checkpoints:
  • Create 'result' folder manually under 'CaDA/50' and 'CaDA/100'.
  • Download checkpoint.zip from Hugging Face.
  • Unzip checkpoint.zip. It will produce two directories: 50 and 100.
    • Inside 50, you will find a folder named 2024-1111-1139.
    • Inside 100, you will find a folder named 2024-1121-1355.
  • Organize them into the project directory as follows:
CaDA
├── data
│   ├── lib_data
│   └── synthetic_data
├── 50
│   └── result
│       └── 2024-1111-1139
├── 100
│   └── result
│       └── 2024-1121-1355
└── utils
  1. Prepare environment:

The project is developed with Python 3.8.15. Key packages include:

torch     2.0.1
torchrl   0.1.1
rl4co     0.2.0
tensordict                   0.1.2

The complete list of dependencies can be found in requirements.txt.

Training and Testing

For detailed instructions on training and testing the model, please refer to the README files inside the 50 and 100 directories.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages