TensorFlow Serving is a flexible, high-performance serving system for machine learning models, designed for production environments. It deals with the inference aspect of machine learning, taking models after training and managing their lifetimes, providing clients with versioned access via a high-performance, reference-counted lookup table. TensorFlow Serving provides out-of-the-box integration with TensorFlow models, but can be easily extended to serve other types of models and data.
See tensorflow.org for more information This image is built by IBM to run on the IBM Z architecture and is not affiliated with any other community that provides a version of this image.2.4.0 | docker pull icr.io/ibmz/tensorflow-serving@sha256:d232a0532342a29ed49d9cd61957793af07da6e8fba4d4c1da808124bb5909b7 | Vulnerability Report | 07-29-2021 | 2.4-opts-vector | docker pull icr.io/ibmz/tensorflow-serving@sha256:c1a9fb946305538b6c72967cf349c9504f08c2054dea4fd575fd37078bfd1d1d | Vulnerability Report | 08-09-2022 | 2.7.0 | docker pull icr.io/ibmz/tensorflow-serving@sha256:8da2e8e497fc839a76cad33b16a76e1ed537730b762a4c7f17fb2673e27fcf55 | Vulnerability Report | 05-04-2022 |
Version | Pull String | Security (IBM Cloud) | Created |
---|---|---|---|
Exposed Ports: 8500
(gRPC), 8501
(REST)
docker pull icr.io/ibmz/tensorflow-serving:[version&hash]
git clone -b 2.4.0 https://github.com/tensorflow/serving
Run a Base Container:
docker run -d --name serving_base icr.io/ibmz/tensorflow-serving:[version&hash]
Copy Your Model Data:
docker cp /path/to/your/model_data_from_git_clone_step serving_base:/models/half_plus_two
Create a Custom Container and set MODEL_NAME:
docker commit --change "ENV MODEL_NAME=half_plus_two" serving_base serving_custom
docker stop serving_base
docker rm serving_base
Run TensorFlow Serving Container:
docker run -t --rm -p 8501:8501 serving_custom
Query the model using the predict API
curl -d '{"instances": [1.0, 2.0, 5.0]}' \
-X POST http://[host_machine]:8501/v1/models/half_plus_two:predict
Returns => { "predictions": [2.5, 3.0, 4.5] }