Icon

4. Retrain Model

4. Retrain Model
KAFKA CONSUMER (to "fraud-response" topic)- kafka consumer consume data from topic "fraud-response" and update it in the database- the updated database will be the new input data for model re-training MODEL RETRAIN- after a certain of number of additional data, the model need to be retrained to keep the model performance and effectivness.- the input data will be the previous raw data + new additional data- the previous trained model will be re-trained- the re-trained model is saved for model deployment again re-train the modelupload trained modelconvert data to3 dimensional arrayConnect tokafka containerConsumetopic" fraud-response" from kafkaConvert kafka "fraud-response" to tablesplit train:validation(80:20)Evaluate the modelsave re-trained modelNode 56rename "partitions"column to "fraud"Insert data to tablein database Keras NetworkLearner Keras NetworkReader Train DataTransformation Kafka Connector Kafka Consumer JSON to Table Partitioning Keras NetworkExecutor Keras NetworkWriter MySQL Connector Column Rename DB Insert Rule Engine Joiner Scorer KAFKA CONSUMER (to "fraud-response" topic)- kafka consumer consume data from topic "fraud-response" and update it in the database- the updated database will be the new input data for model re-training MODEL RETRAIN- after a certain of number of additional data, the model need to be retrained to keep the model performance and effectivness.- the input data will be the previous raw data + new additional data- the previous trained model will be re-trained- the re-trained model is saved for model deployment again re-train the modelupload trained modelconvert data to3 dimensional arrayConnect tokafka containerConsumetopic" fraud-response" from kafkaConvert kafka "fraud-response" to tablesplit train:validation(80:20)Evaluate the modelsave re-trained modelNode 56rename "partitions"column to "fraud"Insert data to tablein database Keras NetworkLearner Keras NetworkReader Train DataTransformation Kafka Connector Kafka Consumer JSON to Table Partitioning Keras NetworkExecutor Keras NetworkWriter MySQL Connector Column Rename DB Insert Rule Engine Joiner Scorer

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: suhaimi
Created at: 2022-06-27
On NodePit since: 2022-08-21
Last update: 2024-04-24
Created with KNIME version: v4.6.1
Tags: model trainingdeep learningclassification

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!