Share via


SdcaMulticlassTrainerBase<TModel> Class

Definition

The IEstimator<TTransformer> to predict a target using a linear multiclass classifier model trained with a coordinate descent method. Depending on the used loss function, the trained model can be, for example, maximum entropy classifier or multi-class support vector machine.

public abstract class SdcaMulticlassTrainerBase<TModel> : Microsoft.ML.Trainers.SdcaTrainerBase<Microsoft.ML.Trainers.SdcaMulticlassTrainerBase<TModel>.MulticlassOptions,Microsoft.ML.Data.MulticlassPredictionTransformer<TModel>,TModel> where TModel : class
type SdcaMulticlassTrainerBase<'Model (requires 'Model : null)> = class
    inherit SdcaTrainerBase<SdcaMulticlassTrainerBase<'Model>.MulticlassOptions, MulticlassPredictionTransformer<'Model>, 'Model (requires 'Model : null)>
Public MustInherit Class SdcaMulticlassTrainerBase(Of TModel)
Inherits SdcaTrainerBase(Of SdcaMulticlassTrainerBase(Of TModel).MulticlassOptions, MulticlassPredictionTransformer(Of TModel), TModel)

Type Parameters

TModel
Inheritance
Derived

Remarks

To create this trainer for maximum entropy classifier, use SdcaMaximumEntropy or SdcaMaximumEntropy(Options). To create this trainer for a loss function (such as support vector machine's hinge loss) of your choice, use SdcaNonCalibrated or SdcaNonCalibrated(Options).

Input and Output Columns

The input label column data must be key type and the feature column must be a known-sized vector of Single.

This trainer outputs the following columns:

Output Column Name Column Type Description
Score Vector of Single The scores of all classes. Higher value means higher probability to fall into the associated class. If the i-th element has the largest value, the predicted label index would be i. Note that i is zero-based index.
PredictedLabel key type The predicted label's index. If its value is i, the actual label would be the i-th category in the key-valued input label type.

Trainer Characteristics

Machine learning task Multiclass classification
Is normalization required? Yes
Is caching required? No
Required NuGet in addition to Microsoft.ML None
Exportable to ONNX Yes

Scoring Function

This trains linear model to solve multiclass classification problems. Assume that the number of classes is $m$ and number of features is $n$. It assigns the $c$-th class a coefficient vector $\textbf{w}_c \in {\mathbb R}^n$ and a bias $b_c \in {\mathbb R}$, for $c=1,\dots,m$. Given a feature vector $\textbf{x} \in {\mathbb R}^n$, the $c$-th class's score would be $\hat{y}^c = \textbf{w}_c^T \textbf{x} + b_c$. If $\textbf{x}$ belongs to class $c$, then $\hat{y}^c$ should be much larger than 0. In contrast, a $\hat{y}^c$ much smaller than 0 means the desired label should not be $c$.

If and only if the trained model is a maximum entropy classifier, you can interpret the output score vector as the predicted class probabilities because softmax function may be applied to post-process all classes' scores. More specifically, the probability of $\textbf{x}$ belonging to class $c$ is computed by $\tilde{P}( c | \textbf{x} ) = \frac{ e^{\hat{y}^c} }{ \sum_{c' = 1}^m e^{\hat{y}^{c'}} }$ and store at the $c$-th element in the score vector. In other cases, the output score vector is just $[\hat{y}^1, \dots, \hat{y}^m]$.

Training Algorithm Details

The optimization algorithm is an extension of a coordinate descent method following a similar path proposed in an earlier paper. It is usually much faster than L-BFGS and truncated Newton methods for large-scale and sparse data sets.

This class uses empirical risk minimization (i.e., ERM) to formulate the optimization problem built upon collected data. Note that empirical risk is usually measured by applying a loss function on the model's predictions on collected data points. If the training data does not contain enough data points (for example, to train a linear model in $n$-dimensional space, we need at least $n$ data points), overfitting may happen so that the model produced by ERM is good at describing training data but may fail to predict correct results in unseen events. Regularization is a common technique to alleviate such a phenomenon by penalizing the magnitude (usually measured by the norm function) of model parameters. This trainer supports elastic net regularization, which penalizes a linear combination of L1-norm (LASSO), $|| \textbf{w}_c ||_1$, and L2-norm (ridge), $|| \textbf{w}_c ||_2^2$ regularizations for $c=1,\dots,m$. L1-norm and L2-norm regularizations have different effects and uses that are complementary in certain respects.

Together with the implemented optimization algorithm, L1-norm regularization can increase the sparsity of the model weights, $\textbf{w}_1,\dots,\textbf{w}_m$. For high-dimensional and sparse data sets, if users carefully select the coefficient of L1-norm, it is possible to achieve a good prediction quality with a model that has only a few non-zero weights (e.g., 1% of total model weights) without affecting its prediction power. In contrast, L2-norm cannot increase the sparsity of the trained model but can still prevent overfitting by avoiding large parameter values. Sometimes, using L2-norm leads to a better prediction quality, so users may still want to try it and fine tune the coefficients of L1-norm and L2-norm. Note that conceptually, using L1-norm implies that the distribution of all model parameters is a Laplace distribution while L2-norm implies a Gaussian distribution for them.

An aggressive regularization (that is, assigning large coefficients to L1-norm or L2-norm regularization terms) can harm predictive capacity by excluding important variables from the model. For example, a very large L1-norm coefficient may force all parameters to be zeros and lead to a trivial model. Therefore, choosing the right regularization coefficients is important in practice.

Check the See Also section for links to usage examples.

Fields

FeatureColumn

The feature column that the trainer expects.

(Inherited from TrainerEstimatorBase<TTransformer,TModel>)
LabelColumn

The label column that the trainer expects. Can be null, which indicates that label is not used for training.

(Inherited from TrainerEstimatorBase<TTransformer,TModel>)
WeightColumn

The weight column that the trainer expects. Can be null, which indicates that weight is not used for training.

(Inherited from TrainerEstimatorBase<TTransformer,TModel>)

Properties

Info (Inherited from StochasticTrainerBase<TTransformer,TModel>)

Methods

Fit(IDataView)

Trains and returns a ITransformer.

(Inherited from TrainerEstimatorBase<TTransformer,TModel>)
GetOutputSchema(SchemaShape) (Inherited from TrainerEstimatorBase<TTransformer,TModel>)

Extension Methods

AppendCacheCheckpoint<TTrans>(IEstimator<TTrans>, IHostEnvironment)

Append a 'caching checkpoint' to the estimator chain. This will ensure that the downstream estimators will be trained against cached data. It is helpful to have a caching checkpoint before trainers that take multiple data passes.

WithOnFitDelegate<TTransformer>(IEstimator<TTransformer>, Action<TTransformer>)

Given an estimator, return a wrapping object that will call a delegate once Fit(IDataView) is called. It is often important for an estimator to return information about what was fit, which is why the Fit(IDataView) method returns a specifically typed object, rather than just a general ITransformer. However, at the same time, IEstimator<TTransformer> are often formed into pipelines with many objects, so we may need to build a chain of estimators via EstimatorChain<TLastTransformer> where the estimator for which we want to get the transformer is buried somewhere in this chain. For that scenario, we can through this method attach a delegate that will be called once fit is called.

Applies to

See also