license: mit
tags:
- Astronomy
- Classification
- Object Detection
DRAFTS
✨ Deep learning-based RAdio Fast Transient Search pipeline✨
Description
Here is the model repository for the Deep learning-based RAdio Fast Transient Search pipeline (DRAFTS).
We invite you to stay tuned for updates on the remaining components and different versions of the models.
Usage
There are four .pth files in this repository.
Object Detection
The files starting with cent are the trained model checkpoints for the object detection models in the DRAFTS project.
To load the object detection model
import torch
from centernet_model import centernet
base_model = 'resnet18' # 'resnet50'
model = centernet(model_name=base_model)
model.load_state_dict(torch.load('cent_{}.pth'.format(base_model)))
model.eval()
Binary Classification
The files starting with class are the trained model checkpoints for the classification models in the DRAFTS project.
To load the classification detection model
import torch
from binary_model import BinaryNet
base_model = 'resnet18' # 'resnet50'
model = BinaryNet(base_model, num_classes=2)
model.load_state_dict(torch.load('class_{}.pth'.format(base_model)))
model.eval()
Contributing
We welcome contributions to the DRAFTS project! If you have any suggestions, bug reports, or feature requests, please open an issue or submit a pull request on our GitHub repository.
Contact
For any questions or inquiries, please contact us at [email protected] or [email protected]