Description
The data_processing pipeline takes in the raw input data Shuttles, Companies and Reviews. Shuttles and Companies require some pre-processing (converting strings to float/boolean values)
before the 3 input tables are merged into a single model_input_table to be used in model creation.
The data_science pipeline consumes the model_input_table and generates models from it. There are 2 variants, one is the active_modelling_pipeline and the other is a candidate_modelling_pipeline demonstrating the use of modular pipelines and how to use the same code with different sets of parameters.
Automatic Doc creation
This is a standalone KedroViz React component in a static website. It can be embedded in the document as seen here or accessed as a separate page.