train_test_assign: Training and testing split

In metasnf: Meta Clustering with Similarity Network Fusion

Training and testing split

Description

Given a vector of uid_id and a threshold, returns a list of which members should be in the training set and which should be in the testing set. The function relies on whether or not the absolute value of the Jenkins's one_at_a_time hash function exceeds the maximum possible value (2147483647) multiplied by the threshold.

Usage

train_test_assign(train_frac, uids, seed = 42)

Arguments

train_frac	The fraction (0 to 1) of observations for training
uids	A character vector of UIDs to be distributed into training and test sets.
seed	Seed used for Jenkins's one_at_a_time hash function.

Value

A named list containing the training and testing uid_ids.

metasnf documentation built on April 3, 2025, 5:40 p.m.