Skip to main content

Run a NERDD module as a service

Project description

NERDD-Link

Run a NERDD module as a
service that consumes input molecules and produces prediction tuples.

Installation

pip install -U nerdd-link

Usage

When a class inherits from nerdd_module.AbstractModel (see NERDD Module Github page), it can be used to create a Kafka service.

# run a Kafka service for NerddModel on localhost:9092
run_nerdd_server package.path.to.NerddModel

# modify broker url, input topic and batch size
run_nerdd_server package.path.to.NerddModel \
  --broker-url my-cluster-kafka-bootstrap.kafka:9092 \
  --input-topic examples \
  --batch-size 10

# more information via --help
run_nerdd_server --help

If the model class is called ExamplePredictionModel, the server will read input tuples from the input topic example-prediction-inputs in batches of size 100 and write results to the results topic. The batch size specifies the number of input tuples that are given to the model at once.

Communication

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

nerdd_link-0.4.0.tar.gz (17.0 kB view details)

Uploaded Source

Built Distribution

nerdd_link-0.4.0-py3-none-any.whl (19.9 kB view details)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page