RegressionConfig
Common SDCA regression settings.
Record fields
| Name | Description |
|---|---|
L1Regularization | L1 regularization weight, or None to use the ML.NET default. |
L2Regularization | L2 regularization weight, or None to use the ML.NET default. |
MaximumNumberOfIterations | Maximum training iterations, or None to use the ML.NET default. |
Seed | ML.NET context seed. The default is Some 0 for reproducibility. |
RegressionConfig module
Constructors for SDCA regression settings.
| Function | Description |
|---|---|
create () | Create settings from the ML.NET SDCA defaults with a deterministic context seed. |
create
create ()Create settings from the ML.NET SDCA defaults with a deterministic context seed.
Parameters
():unit
Returns RegressionConfig
RegressionModel
A fitted SDCA regression model.
Instance members
| Name | Description |
|---|---|
this.FeatureCount | Number of input features expected by the model. |
this.Transformer | Underlying ML.NET model. |
Regression
SDCA regression operations.
| Function | Description |
|---|---|
evaluate dataset model | Evaluate a fitted regressor with ML.NET regression metrics. |
fit config dataset | Fit an SDCA regressor with common F# settings. |
fitWithOptions seed configure dataset | Fit an SDCA regressor after configuring the complete ML.NET options object. |
load path | Load a linear regression model compatible with the SDCA output type. |
predict features model | Predict target values. Values are returned as a detached CPU float32 tensor. |
save path model | Save a fitted model in the ML.NET zip format. |
evaluate
evaluate dataset modelEvaluate a fitted regressor with ML.NET regression metrics.
Parameters
dataset:RegressionDatasetmodel:RegressionModel
Returns RegressionMetrics
fit
fit config datasetFit an SDCA regressor with common F# settings.
Parameters
config:RegressionConfigdataset:RegressionDataset
Returns RegressionModel
fitWithOptions
fitWithOptions seed configure datasetFit an SDCA regressor after configuring the complete ML.NET options object.
Parameters
seed:int optionconfigure:Options -> unitdataset:RegressionDataset
Returns RegressionModel
load
load pathLoad a linear regression model compatible with the SDCA output type.
Parameters
path:string
Returns RegressionModel
predict
predict features modelPredict target values. Values are returned as a detached CPU float32 tensor.
Parameters
features:Tensormodel:RegressionModel
Returns Tensor
save
save path modelSave a fitted model in the ML.NET zip format.
Parameters
path:stringmodel:RegressionModel