Briko's GameNMT Demo

Briko's GameNMT Demo can be found at: https://www.briko.org/demo

If you'd like to try the pre-trained model at a local environment for a test yourself, here is a simple tutorial you can follow:

1.Download this repository:

$ git clone git@github.com:briko-org/GameNMT-demo.git

It contains the code you need to run the pre-trained transformer model.
For demonstration purpose, we assume that you have download GameNMT-demo project to your $HOME directory.

2.Install Python (3.6):

$ sudo add-apt-repository ppa:jonathonf/python-3.6
$ sudo apt-get update
$ sudo apt-get install python3.6

3.Install TensorFlow (TensorFlow 1.13.1)

CPU-only:

$ pip3 install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.13.1-cp36-cp36m-linux_x86_64.whl

GPU-only:

$ pip3 install https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.13.1-cp36-cp36m-linux_x86_64.whl

4.Add the GameNMT-demo/models folder to the Python path with the command:

$ export PYTHONPATH="$PYTHONPATH:$HOME/GameNMT-demo/models"

5.Install dependencies:

$ pip3 install --user -r $HOME/GameNMT-demo/models/official/requirements.txt

6.Export variables

PARAM_SET=base
DATA_DIR=$HOME/GameNMT-demo/models/official/transformer/data
MODEL_DIR=$HOME/GameNMT-demo/models/official/transformer/model_$PARAM_SET
VOCAB_FILE=$DATA_DIR/vocab.file

7.Download the trained model to MODEL_DIR:

Download Torrent

8.Finally, you can try to translate some text using the trained model with the command below:

$ python3.6 $HOME/GameNMT-demo/models/official/transformer/translate.py --model_dir=$MODEL_DIR --vocab_file=$VOCAB_FILE \
    --param_set=$PARAM_SET --text="hello world"