Expand description
Online models: baseline regressors, linear regression, logistic regression, Naive Bayes classifiers, and FTRL-Proximal for sparse features.
Structs§
- Baseline
Config - Configuration shared by baseline regressors.
- Bernoulli
Naive Bayes - Online Bernoulli Naive Bayes classifier.
- Exponentially
Weighted Mean Regressor - A regressor that predicts an exponentially weighted mean of targets.
- Ftrl
Classifier - FTRL binary classifier with log loss.
- Ftrl
Config - Configuration for FTRL models.
- Ftrl
Regressor - FTRL regressor with squared loss.
- Ftrl
Resource Diagnostics - Bounded-resource diagnostics for a sparse FTRL model.
- Gaussian
Naive Bayes - Online Gaussian Naive Bayes classifier.
- Last
Value Regressor - A regressor that always predicts the last observed target.
- Linear
Regression - Online linear regression model.
- Linear
Regression Config - Configuration for
LinearRegression. - Logistic
Regression - Online logistic regression model.
- Logistic
Regression Config - Configuration for
LogisticRegression. - Mean
Regressor - A regressor that always predicts the running mean of observed targets.
- Multinomial
Naive Bayes - Online Multinomial Naive Bayes classifier.
- Naive
Bayes Config - Configuration for Naive Bayes classifiers.
Enums§
- NewFeature
Policy - Policy for handling new
FeatureIds onceFtrlConfig::max_featureshas been reached.