Accessing the files

The IBM Entitled Registry provides access to the IBM Watson Natural Language Processing Library for Embed images. Once you've obtained the entitlement key from the container software library, you can login to the registry with the key, and pull the images to your local machine.

Prerequisites

Docker version 18.06.0 or greater

Getting the Docker images

Once you have obtained your IBM Entitled Registry entitlement key, run the following command to allow Docker to access the images.

docker login cp.icr.io --username cp --password <entitlement_key>

Watson Natural Language Processing Library for Embed includes various container images for working with Watson NLP.

Watson NLP Runtime image

The list of available Watson NLP Runtime images are:

cp.icr.io/cp/ai/watson-nlp-runtime:1.0.18
cp.icr.io/cp/ai/watson-nlp-runtime:1.0.19
cp.icr.io/cp/ai/watson-nlp-runtime:1.0.20
cp.icr.io/cp/ai/watson-nlp-runtime:1.0.30
cp.icr.io/cp/ai/watson-nlp-runtime:1.1.0
cp.icr.io/cp/ai/watson-nlp-runtime:1.1.15
cp.icr.io/cp/ai/watson-nlp-runtime:1.1.26
cp.icr.io/cp/ai/watson-nlp-runtime:1.1.30
cp.icr.io/cp/ai/watson-nlp-runtime:1.1.32
cp.icr.io/cp/ai/watson-nlp-runtime:1.1.36

Note: Each NLP Runtime image is released with multiple tags, for example, along with the 1.0.30 image tag, we also push a 1.0 and 1 tag. This is done so that users can subscribe to image updates within a major (1.x.x) or minor (1.0.x) stream.

The Watson Runtime on its own doesn't have any models included. However, you can easily build a runtime container image to include one or more pretrained models

Pretrained model images

Pretrained models are also stored as container images in the IBM Entitled Registry. The available model images are listed in the models catalog section.

Next steps

Deploy the runtime container image using one of several available options. See Running the containers for details.