API Reference¶
This is the API documentation for scikit-multiflow
.
Core¶
The skmultiflow.core
module covers core elements of scikit-multiflow.
Base Estimator class for compatibility with scikit-learn. |
|
Base class for most objects in scikit-multiflow |
|
Mixin class for all classifiers in scikit-multiflow. |
|
Mixin class for all regression estimators in scikit-multiflow. |
|
Mixin class for all meta estimators in scikit-multiflow. |
|
Mixin to mark estimators that support multioutput. |
|
[Experimental] Holds a set of sequential operation (transforms), followed by a single estimator. |
Data¶
The skmultiflow.data
module contains data stream methods including methods for
batch-to-stream conversion and generators.
Base Stream class. |
|
Creates a stream from a data source. |
|
Creates a stream from a file source. |
|
Generates a stream with concept drift. |
|
Create a temporal stream from a data source. |
Stream Generators¶
Agrawal stream generator. |
|
Simulate a stream with anomalies in sine waves |
|
Hyperplane stream generator. |
|
LED stream generator. |
|
LED stream generator with concept drift. |
|
Mixed data stream generator. |
|
Random Radial Basis Function stream generator. |
|
Random Radial Basis Function stream generator with concept drift. |
|
Random Tree stream generator. |
|
SEA stream generator. |
|
Sine stream generator. |
|
STAGGER concepts stream generator. |
|
Waveform stream generator. |
|
Creates a multi-label stream. |
|
Creates a regression stream. |
Learning methods¶
Anomaly detection methods¶
The skmultiflow.anomaly_detection
module includes anomaly detection methods.
Half–Space Trees. |
Bayes methods¶
The skmultiflow.bayes
module includes Bayes learning methods.
Naive Bayes classifier. |
Lazy learning methods¶
The skmultiflow.lazy
module includes lazy learning methods in which generalization of the training
data is delayed until a query is received, this is, on-demand.
k-Nearest Neighbors classifier. |
|
K-Nearest Neighbors classifier with ADWIN change detector. |
|
Self Adjusting Memory coupled with the kNN classifier. |
|
k-Nearest Neighbors regressor. |
Ensemble methods¶
The skmultiflow.meta
module includes meta learning methods.
Accuracy Weighted Ensemble classifier |
|
Adaptive Random Forest classifier. |
|
Adaptive Random Forest regressor. |
|
Additive Expert ensemble classifier. |
|
Batch Incremental ensemble classifier. |
|
Classifier Chains for multi-label learning. |
|
Probabilistic Classifier Chains for multi-label learning. |
|
Monte Carlo Sampling Classifier Chains for multi-label learning. |
|
Dynamic Weighted Majority ensemble classifier. |
|
Learn++.NSE ensemble classifier. |
|
Learn++ ensemble classifier. |
|
Leveraging Bagging ensemble classifier. |
|
Multi-Output Learner for multi-target classification or regression. |
|
Online AdaC2 ensemble classifier. |
|
Online Boosting ensemble classifier. |
|
Online CSB2 ensemble classifier. |
|
Online RUSBoost ensemble classifier. |
|
Online SMOTEBagging ensemble classifier. |
|
Online Under-Over-Bagging ensemble classifier. |
|
Oza Bagging ensemble classifier. |
|
Oza Bagging ensemble classifier with ADWIN change detector. |
|
Regressor Chains for multi-output learning. |
|
Streaming Random Patches ensemble classifier. |
Neural Networks¶
The skmultiflow.neural_networks
module includes learning methods based on Neural Networks.
Mask for sklearn.linear_model.Perceptron. |
Prototype based methods¶
The skmultiflow.prototype
module includes prototype-based learning methods.
Robust Soft Learning Vector Quantization for Streaming and Non-Streaming Data. |
Rules based methods¶
The skmultiflow.rules
module includes rule-based learning methods.
Very Fast Decision Rules classifier. |
Trees based methods¶
The skmultiflow.trees
module includes learning methods based on trees.
Hoeffding Tree or Very Fast Decision Tree classifier. |
|
Hoeffding Adaptive Tree classifier. |
|
Extremely Fast Decision Tree classifier. |
|
Label Combination Hoeffding Tree for multi-label classification. |
|
Hoeffding Tree regressor. |
|
Hoeffding Adaptive Tree regressor. |
|
Incremental Structured Output Prediction Tree (iSOUP-Tree) for multi-target regression. |
|
Stacked Single-target Hoeffding Tree regressor. |
Drift Detection¶
The skmultiflow.drift_detection
module includes methods for Concept Drift Detection.
Adaptive Windowing method for concept drift detection. |
|
Drift Detection Method. |
|
Early Drift Detection Method. |
|
Drift Detection Method based on Hoeffding’s bounds with moving average-test. |
|
Drift Detection Method based on Hoeffding’s bounds with moving weighted average-test. |
|
Kolmogorov-Smirnov Windowing method for concept drift detection. |
|
Page-Hinkley method for concept drift detection. |
Evaluation¶
The skmultiflow.evaluation
module includes evaluation methods for stream learning.
The holdout evaluation method or periodic holdout evaluation method. |
|
The prequential evaluation method or interleaved test-then-train method. |
|
The prequential evaluation delayed method. |
Transform¶
The skmultiflow.transform
module covers methods that perform data transformations.
Fill missing values with some defined value. |
|
Transform one-hot encoded data into categorical feature(s). |
|
Transform features by scaling each feature to a given range. |
|
Standardize features by removing the mean and scaling to unit variance. |