init(convolutionWeightsWithInputFeatureChannelCount:outputFeatureChannelCount:dataType:)
Creates a tensor descriptor with the number of feature channels and data type you specify.
Declaration
convenience init?(convolutionWeightsWithInputFeatureChannelCount inputFeatureChannelCount: Int, outputFeatureChannelCount: Int, dataType: MLCDataType)Parameters
- inputFeatureChannelCount:
The number of input feature channels.
- outputFeatureChannelCount:
The number of output feature channels.
- dataType:
The tensor data type.
Discussion
This initializer provides a convenient way for you to configure weights tensors with a kernel of size 1.
See Also
Creating Tensor Descriptors
init(shape:dataType:)init(shape:sequenceLengths:sortedSequences:dataType:)init(width:height:featureChannelCount:batchSize:)init(width:height:featureChannelCount:batchSize:dataType:)init(convolutionWeightsWithWidth:height:inputFeatureChannelCount:outputFeatureChannelCount:dataType:)init(convolutionBiasesWithFeatureChannelCount:dataType:)maxTensorDimensions