Training Pipeline

Created: 2023-02-13 14:32
#note

A robust managed training pipeline helps to create repeatable ML training and testing workflows while reducing human costs -> useful for traceability, reproducibility, and automation.

Functions:

• get training dataset from feature store;
• get parameters for training;
• build and validate model;
• get test data from feature store;
• test the model and review results;
• update parameters of the model.

Tags

#mlops #ml #training