Implementation:Online ml River Stream Iter Polars

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

Overview

Converts Polars DataFrames into River-compatible data streams for online learning.

Description

The iter_polars function enables streaming over rows of Polars DataFrames, a fast DataFrame library for Python. It converts DataFrame rows into dictionary format suitable for River's online learning algorithms. The function supports both Series and DataFrame targets, including multi-output scenarios.

Usage

Use this when working with Polars DataFrames and want to apply River's online learning algorithms. It's particularly useful when you have data stored or processed in Polars format and need to convert it to River's streaming interface.

Code Reference

Source Location

• Repository: Online_ml_River
• File: river/stream/iter_polars.py

Signature

def iter_polars(
    X: pl.DataFrame,
    y: pl.Series | pl.DataFrame | None = None,
    **kwargs
) -> base.typing.Stream:
    ...

Import

from river import stream

I/O Contract

Parameter	Type	Description
X	pl.DataFrame	Polars DataFrame containing features
y	pl.Series, pl.DataFrame, or None	Target variable(s) as Series or DataFrame
**kwargs	dict	Additional arguments passed to stream.iter_array

Returns:

Type	Description
Iterator[(dict, Any)]	Stream of (features dict, target) tuples

Usage Examples

import polars as pl
from river import stream

# Create a Polars DataFrame
X = pl.DataFrame({
    'x1': [1, 2, 3, 4],
    'x2': ['blue', 'yellow', 'yellow', 'blue'],
    'y': [True, False, False, True]
})

# Extract target column
y = X.get_column('y')
X = X.drop("y")

# Iterate over the data
for features, target in stream.iter_polars(X, y):
    print(f"Features: {features}")
    print(f"Target: {target}")
    print()

# Example with multi-output
X_multi = pl.DataFrame({
    'feature1': [1.0, 2.0, 3.0],
    'feature2': [4.0, 5.0, 6.0]
})

y_multi = pl.DataFrame({
    'target1': [0.1, 0.2, 0.3],
    'target2': [0.4, 0.5, 0.6]
})

print("Multi-output example:")
for x, y_dict in stream.iter_polars(X_multi, y_multi):
    print(f"Features: {x}")
    print(f"Targets: {y_dict}")
    print()

# With shuffling (passed through kwargs)
shuffled_stream = stream.iter_polars(
    X, y,
    shuffle=True,
    seed=42
)

print("Shuffled data:")
for x, y_val in shuffled_stream:
    print(f"{x} -> {y_val}")

Related Pages

• Environment:Online_ml_River_Python_Runtime_Environment
• Online_ml_River_Stream_Iter_Pandas
• Online_ml_River_Stream_Iter_Array