Burges, Christopher, et al. "Learning to rank using gradient descent." Proceedings of the 22nd International Conference on Machine learning (ICML-05). 2005.
pytorch, pytorch-ignite, torchviz, numpy tqdm matplotlib
pytorch: see the official document.
$ pip install pytorch-ignite torchviz numpy tqdm matplotlib
chainer, matplotlib, numpy, tqdm
$ pip install chainer matplotlib numpy tqdm
- Train a ranking model
$ python train.py
-h
option shows help.
$ python train.py -h
usage: train.py [-h] [-b BATCH_SIZE] [-e EPOCH]
trains a ranking model for mnist
optional arguments:
-h, --help show this help message and exit
-b BATCH_SIZE, --batch_size BATCH_SIZE
batch size
-e EPOCH, --epoch EPOCH
epoch
- Visualize scores for test data
$ python visualize.py -m model_file -o output_file
-h
option shows help.
$ python visualize.py -h
usage: visualize.py [-h] -m M [-b B] [-o O] [-t T]
visualizes scores for test dataset
optional arguments:
-h, --help show this help message and exit
-m M model file generated from train.py
-b B batch size
-o O output file
-t T title of the figure