Concepts¶
Here we briefly explain some of the concepts and terminology used within the Draco project and documentation.
Primitive¶
We call the smallest computational blocks used in a Machine Learning process primitives, which:
Can be either classes or functions.
Have some initialization arguments, which MLBlocks calls
init_params.Have some tunable hyperparameters, which have types and a list or range of valid values.
Template¶
Primitives can be combined to form what we call Templates, which:
Have a list of primitives.
Have some initialization arguments, which correspond to the initialization arguments of their primitives.
Have some tunable hyperparameters, which correspond to the tunable hyperparameters of their primitives.
Pipeline¶
Templates can be used to build Pipelines by taking and fixing a set of valid hyperparameters for a Template. Hence, Pipelines:
Have a list of primitives, which corresponds to the list of primitives of their template.
Have some initialization arguments, which correspond to the initialization arguments of their template.
Have some hyperparameter values, which fall within the ranges of valid tunable hyperparameters of their template.
A pipeline can be fitted and evaluated directly using MLBlocks, or using the DracoPipeline.
Tuning¶
We call tuning the process of, given a dataset and a collection of templates, finding the pipeline derived from the templates that gets the best possible score on the dataset.
This process usually involves fitting and evaluating multiple pipelines with different hyperparameter configurations on the same data while using optimization algorithms to deduce which hyperparameters are more likely to get the best results in the next iterations.
We call each one of these evaluations a tuning iteration.
The process of selecting and tuning the templates is handled by a BTBSession, which is responsible for discarding the templates that do not work on the given data and for keeping track of the template and hyperparameters that obtain the best performance.
DracoPipeline¶
This class is the one in charge of loading the MLBlocks Pipelines configured in the system and use them to learn from the data and make predictions.
This class is also responsible for creating the BTBSession that will handle the selection and tuning of the templates.