NeuralTQE
A stacking model of quality estimation implemented with PyTorch
Training and Testing Data Format
Training and Testing data should be in the format as : source_sentence \t target_sentence \t \ score Note that all sentences should be segmented.
Using GPU
If you are running the code with GPU, simply set the use_cuda=torch.cuda.is_available(), otherwise set use_cuda=False
Pretrained Embeddings
Download or pretrain monolingual or crosslingual embeddings and have the first line in each file deleted. Note that pretrained embeddings may be problematic due to some lines contain empty words or null vectors.
Run the Code
To run the code,
- configure your parameters in the mtmain.py file.
- run
python mtmain.py.