House Price Regression¶
House Price Regression refers to the prediction of house prices based on various factors, using the XGBoost Regression model (in our case). In this example, we will train our data on the XGBoost model to predict house prices in multiple regions.
Where Does Flyte Fit In?¶
Orchestrates the machine learning pipeline.
Helps cache the output state between
tasks.Easier backtracking to the error source.
Provides a Rich UI to view and manage the pipeline.
House price prediction pipeline for one region doesn’t require a dynamic() workflow. When multiple regions are involved, to iterate through the regions at run-time and thereby build the DAG, Flyte workflow has to be dynamic().
Tip
Refer to Dynamic workflows section to learn more about dynamic workflows.
Dataset¶
We will create a custom dataset to build our model by referring to the SageMaker example.
The dataset will have the following columns:
Price
House Size
Number of Bedrooms
Year Built
Number of Bathrooms
Number of Garage Spaces
Lot Size
Takeaways¶
An in-depth dive into dynamic workflows
How the Flyte type-system works