Dela via


ResNet18Extension.ResNet18 Method

Definition

Overloads

ResNet18(DnnImageModelSelector, IHostEnvironment, String, String)

Returns an estimator chain with the two corresponding models (a preprocessing one and a main one) required for the ResNet pipeline. Also includes the renaming ColumnsCopyingTransforms required to be able to use arbitrary input and output column names. This assumes both of the models are in the same location as the file containing this method, which they will be if used through the NuGet. This should be the default way to use ResNet18 if importing the model from a NuGet.

ResNet18(DnnImageModelSelector, IHostEnvironment, String, String, String)

This allows a custom model location to be specified. This is useful is a custom model is specified, or if the model is desired to be placed or shipped separately in a different folder from the main application. Note that because ONNX models must be in a directory all by themselves for the OnnxTransformer to work, this method appends a ResNet18Onnx/ResNetPrepOnnx subdirectory to the passed in directory to prevent having to make that directory manually each time.

ResNet18(DnnImageModelSelector, IHostEnvironment, String, String)

Returns an estimator chain with the two corresponding models (a preprocessing one and a main one) required for the ResNet pipeline. Also includes the renaming ColumnsCopyingTransforms required to be able to use arbitrary input and output column names. This assumes both of the models are in the same location as the file containing this method, which they will be if used through the NuGet. This should be the default way to use ResNet18 if importing the model from a NuGet.

public static Microsoft.ML.Data.EstimatorChain<Microsoft.ML.Transforms.ColumnCopyingTransformer> ResNet18 (this Microsoft.ML.Transforms.Onnx.DnnImageModelSelector dnnModelContext, Microsoft.ML.Runtime.IHostEnvironment env, string outputColumnName, string inputColumnName);
static member ResNet18 : Microsoft.ML.Transforms.Onnx.DnnImageModelSelector * Microsoft.ML.Runtime.IHostEnvironment * string * string -> Microsoft.ML.Data.EstimatorChain<Microsoft.ML.Transforms.ColumnCopyingTransformer>
<Extension()>
Public Function ResNet18 (dnnModelContext As DnnImageModelSelector, env As IHostEnvironment, outputColumnName As String, inputColumnName As String) As EstimatorChain(Of ColumnCopyingTransformer)

Parameters

dnnModelContext
DnnImageModelSelector
outputColumnName
String
inputColumnName
String

Returns

Applies to

ResNet18(DnnImageModelSelector, IHostEnvironment, String, String, String)

This allows a custom model location to be specified. This is useful is a custom model is specified, or if the model is desired to be placed or shipped separately in a different folder from the main application. Note that because ONNX models must be in a directory all by themselves for the OnnxTransformer to work, this method appends a ResNet18Onnx/ResNetPrepOnnx subdirectory to the passed in directory to prevent having to make that directory manually each time.

public static Microsoft.ML.Data.EstimatorChain<Microsoft.ML.Transforms.ColumnCopyingTransformer> ResNet18 (this Microsoft.ML.Transforms.Onnx.DnnImageModelSelector dnnModelContext, Microsoft.ML.Runtime.IHostEnvironment env, string outputColumnName, string inputColumnName, string modelDir);
static member ResNet18 : Microsoft.ML.Transforms.Onnx.DnnImageModelSelector * Microsoft.ML.Runtime.IHostEnvironment * string * string * string -> Microsoft.ML.Data.EstimatorChain<Microsoft.ML.Transforms.ColumnCopyingTransformer>
<Extension()>
Public Function ResNet18 (dnnModelContext As DnnImageModelSelector, env As IHostEnvironment, outputColumnName As String, inputColumnName As String, modelDir As String) As EstimatorChain(Of ColumnCopyingTransformer)

Parameters

dnnModelContext
DnnImageModelSelector
outputColumnName
String
inputColumnName
String
modelDir
String

Returns

Applies to