Implementation:Online ml River Optim Adam

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

Overview

Adam (Adaptive Moment Estimation) is a popular adaptive learning rate optimizer that combines momentum and RMSProp, maintaining both first and second moment estimates.

Description

Adam computes adaptive learning rates for each parameter by maintaining exponentially decaying averages of both past gradients (first moment) and past squared gradients (second moment). These moment estimates are bias-corrected to account for their initialization at zero. The algorithm updates parameters using the bias-corrected first moment divided by the square root of the bias-corrected second moment, plus a small constant for numerical stability. Adam combines the benefits of AdaGrad (adapting learning rates per parameter) and RMSProp (using exponential moving averages), resulting in an optimizer that works well across a wide range of problems with minimal hyperparameter tuning. The implementation supports both dictionary-based weights and numpy vectors.

Usage

Import from river.optim and use as an optimizer in any River model. Adam is a versatile default choice for most online learning problems.

Code Reference

Source Location

• Repository: Online_ml_River
• File: river/optim/adam.py

Signature

class Adam(optim.base.Optimizer):
    def __init__(self, lr=0.1, beta_1=0.9, beta_2=0.999, eps=1e-8) -> None:
        ...

Import

from river import optim

I/O Contract

Inputs

Name	Type	Required	Description
lr	float	No (default=0.1)	Learning rate
beta_1	float	No (default=0.9)	Exponential decay rate for first moment estimates
beta_2	float	No (default=0.999)	Exponential decay rate for second moment estimates
eps	float	No (default=1e-8)	Small constant for numerical stability

Outputs

Name	Type	Description
optimizer	Adam	Configured optimizer instance ready for model training

Usage Examples

from river import datasets
from river import evaluate
from river import linear_model
from river import metrics
from river import optim
from river import preprocessing

# Create Adam optimizer with default parameters
optimizer = optim.Adam()

# Use with a linear model
dataset = datasets.Phishing()
model = (
    preprocessing.StandardScaler() |
    linear_model.LogisticRegression(optimizer)
)
metric = metrics.F1()

# Evaluate
score = evaluate.progressive_val_score(dataset, model, metric)
print(score)  # F1: 86.52%

# Custom parameters
optimizer = optim.Adam(
    lr=0.01,
    beta_1=0.95,
    beta_2=0.999,
    eps=1e-7
)

model = linear_model.LogisticRegression(optimizer)

# Works well as a general-purpose optimizer
from river import linear_model

# For regression
regressor = linear_model.LinearRegression(optimizer=optim.Adam())

# For classification
classifier = linear_model.LogisticRegression(optimizer=optim.Adam())

Related Pages

• Environment:Online_ml_River_Python_Runtime_Environment