torchcmh is a library built on PyTorch for deep learning cross modal hashing.
if you want use the dataset i make, please download mat file and image file by readme file in dataset package.
- create a configuration file as ./script/default_config.yml
training:
method: DCMH
dataName: Mirflickr25K
batchSize: 64
bit: 64
cuda: True
device: 0
datasetPath:
Mirflickr25k:
img_dir: I:\dataset\mirflickr25k\mirflickr
imgList:
tagList:
labelList:
- run ./script/main.py and input configuration file path.
from torchcmh.run import run
if __name__ == '__main__':
run(config_path='default_config.yml')
- create new method file in folder ./torchcmh/training/
- inherit implement TrainBase
- change the method name in config .yml file and run.