morpheus_dfp.modules.dfp_preproc

Functions

dfp_preproc(builder)[source]

This module function consolidates multiple dfp pipeline modules related to inference/training process into a single module.

Parameters

buildermrc.Builder

Pipeline builder instance.

Notes

Configurable parameters:

• cache_dir (str): Directory for caching intermediate results

• timestamp_column_name (str): Name of the column containing timestamps

• pre_filter_options (dict): Options for pre-filtering control messages

• batching_options (dict): Options for batching files

• user_splitting_options (dict): Options for splitting data by user

• supported_loaders (dict): Supported data loaders for different file types

• monitor_options (dict): Options for monitoring throughput