df_validate: Validate two separate datasets and return validation data on...

In BrandonRCopeland/DataScience: Provides Helper Functions for Data Science Analysis

Validate two separate datasets and return validation data on aggregate and id level features

Description

This function takes two datasets (actual and expected) and validates them based on how many features match exact values at the id level as well as an aggregate

Usage

df_validate(actual, expected, features, key, matchTestPct = 0.98,
  meanTestPct = 0.02)

Arguments

actual	a base R dataframe contain actual feature values
expected	a base R dataframe containing expected feature values
features	the features to be validated
key	The name of the primary key for actual and expected
matchTestPct	The percentage of exact, id level matches to pass test 1
meanTestPct	The maximimun percent difference allowed between actual and expected features

Details

NOTE: The columns to be validate must have the same name in each dataset. If categorical features are included, they will be converted to integer values so means can be established. Integer order is determined by an ascending sort of the categorical variable.

Value

A base R dataframe containing one row per feature with match and mean validation data

BrandonRCopeland/DataScience documentation built on Oct. 14, 2023, 9:45 a.m.