cursus.processing¶

Cursus Processing Module

This module provides access to various data processing utilities and processors that can be used in preprocessing, inference, evaluation, and other ML pipeline steps.

The processors are organized by functionality: - Base processor classes and composition utilities - Text processing (tokenization, NLP) - Numerical processing (imputation, binning) - Categorical processing (label encoding) - Domain-specific processors (BSM, risk tables, etc.)

class Processor[source]¶

Bases: ABC

get_name()[source]¶

abstractmethod process(input_text)[source]¶

processor_name: str¶

function_name_list: List[str]¶

class ComposedProcessor(processors)[source]¶

Bases: Processor

process(input_text)[source]¶

class IdentityProcessor[source]¶

Bases: Processor

An identity processor return a copy of input message itself

process(x)[source]¶

Modules

`categorical`	Categorical Processing Module
`dataloaders`
`datasets`	Dataset classes for PyTorch training pipelines.
`numerical`	Numerical Processing Module
`processor_registry`	Processor Registry for Dynamic Pipeline Construction
`processors`
`temporal`	Temporal Processing Module
`text`
`validation`	Field type validation utilities for preprocessing pipelines.