Machine Learning Clustering on Kafka ingest
Fitting and applying a clustering model on a Kafka stream
Motivation
The purpose of this example is provide you with an example workflow showing an online machine learning model fit on the first N records within a stream and subsequently is using this model for prediction and online updates. In this example we are making use of an streaming implementation of K-Means clustering to filter out one of the generated clusters for persistance to the database.
Example
This example closely follows the Kafka Ingest example provided. This example walks through an example use of functionality provided with the SP ML kdb Insights integration by:
- Deploying a sequential K-Means model to be fit on the first 1000 records provided
- Applying the fit model to subsequent data to make predictions
- Filter the stream based on these predictions and only publish data that meets specified conditions.
This is achieved through filtering the incoming Kafka stream for the trades
topic produced using the example Producer. This stream is decoded and windowed before the above analysis is applied.
To run an ingestion pipeline using the kdb Insights Enterprise, first deploy the base system. Next, download the sample Kafka package as described here. This package will need to be modified to deploy a q
and Python
pipeline containing ML functionality as follows:
View the unpacked package.
export PKG=kafka-ingest
kxi package unpack $PKG-$KX_VERSION.kxi
The default pipeline spec is written in q and is located under kafka-ingest/src/ml.q
in the unpacked package.
The Python pipeline spec is located under kafka-ingest/src/ml.py
.
Deploying
Authenticate with kdb Insights Enterprise and deploy the sample package.
Cleaning up resources
Make sure to teardown any previously installed version of the package and clean up resources before deploying a new version.
Once deployed, the pipeline starts to ingest data and write to the database.
Switch the pipeline to use the q ML spec by replacing the values of base
and spec
in kafka-ingest/pipelines/subway.yaml
.
base: q-ml
..
src: src/ml.q
Then deploy the package using the commands below
kxi auth login
kxi pm push $PKG
kxi pm deploy $PKG
Switch the pipeline to use the Python ML spec by replacing the values of base
and spec
in kafka-ingest/pipelines/subway.yaml
.
base: py-ml
..
src: src/ml.py
Then deploy the package using the commands below
kxi auth login
kxi pm push $PKG
kxi pm deploy $PKG
Checking progress
To check progress of the ingest and validate success, you should follow the steps described here.
Teardown
Tear down the package and data using the command below.
kxi pm teardown --rm-data $PKG