This case study explores the basics of attention as applied to English to French Neural Machine Translation. The same dataset that we created in Chapter 7 is used here (we repeat the preprocessing a second time since we use a separate Docker image).
Without GPU (CPU only):
If using a GPU:
The docker images for this case study are located on dockerhub. Running the commands below will automatically download and start a jupyter notebook.
Run the Docker image for CPU only computation:
docker run -p 8888:8888 --rm springernlp/chapter_9nmt:latest
Run the Docker image with GPU access:
docker run --runtime=nvidia -p 8888:8888 --rm springernlp/chapter_9nmt:latest
docker build -t chapter_9nmt:latest .