Implementation:Online ml River Utils Pretty

Knowledge Sources	Online_ml_River
Domains	Online_Learning, Output_Formatting
Last Updated	2026-02-08 16:00 GMT

Overview

Human-readable formatting utilities for displaying data sizes and tables.

Description

Provides humanize_bytes for converting byte counts to readable formats (B, KiB, MiB, etc.) and print_table for creating formatted text tables from headers and columns. Used throughout River for user-friendly output display.

Usage

Use for displaying model sizes, memory usage, or any metrics in human-readable format. Essential for reporting and debugging.

Code Reference

Source Location

Repository: Online_ml_River
File: river/utils/pretty.py

Signature

def humanize_bytes(n_bytes: int) -> str:
    ...

def print_table(
    headers: list[str],
    columns: list[list[str]],
    order: list[int] | None = None,
):
    ...

Import

from river import utils

Usage Examples

from river import utils

# Humanize byte sizes
print(utils.pretty.humanize_bytes(1024))  # "1 KiB"
print(utils.pretty.humanize_bytes(1536))  # "1.5 KiB"
print(utils.pretty.humanize_bytes(1048576))  # "1 MiB"
print(utils.pretty.humanize_bytes(5000000))  # "4.77 MiB"

# Print tables
headers = ['Model', 'Accuracy', 'Time']
columns = [
    ['Model A', 'Model B', 'Model C'],
    ['0.85', '0.87', '0.83'],
    ['1.2s', '2.1s', '0.9s']
]

table = utils.pretty.print_table(headers, columns)
print(table)
# Output:
# Model      Accuracy   Time
# Model A    0.85       1.2s
# Model B    0.87       2.1s
# Model C    0.83       0.9s

# Custom order
table_ordered = utils.pretty.print_table(
    headers, columns, order=[2, 0, 1]  # C, A, B
)
print(table_ordered)

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment