The KyotoEBMT translation framework takes a tree-to-tree approach to Machine Translation, employing syntactic dependency analysis for both source and target languages in an attempt to preserve non-local structure. The effectiveness of our system is maximized with online example matching and a flexible decoder. Evaluation demonstrates BLEU scores competitive with state-of-the-art SMT systems such as Moses.
HOW TO USE IT
To use this image after having pulled it, you do:
docker run -it -p 80:80 -p 7779:7779 -p 7780:7780 -p 12351:12351 fbergeron/kyoto-ebmt
To start the parse server:
To start the translation server:
To start Apache2:
To use it, launch a web browser from your host machine and visit the page:
DOCKER_IP_ADDRESS should be changed according to your docker settings. The IP address is usually shown when you start docker.
No space left on device
As the image is quite big, it's possible that you will end up seeing an error message saying that no space is left on the device. To prevent this from happening, and also for better performance, it's recommended to create a docker container with more CPUs, more RAM, and more disk space. Before doing this, make sure that no important images are inside your container because they will be deleted when we remove the default machine.
To remove the current default machine:
docker-machine rm default
To create the new default machine:
docker-machine create --driver virtualbox --virtualbox-disk-size 40000 --virtualbox-cpu-count 4 --virtualbox-memory 2000 default