--- title: BNNSNDArrayDescriptor framework: accelerate role: symbol role_heading: Structure path: accelerate/bnnsndarraydescriptor --- # BNNSNDArrayDescriptor A structure that describes the shape, stride, data type, and, optionally, the memory location of an n-dimensional array. ## Declaration ```swift struct BNNSNDArrayDescriptor ``` ## Overview Overview You use a BNNSNDArrayDescriptor structure as the primary mechanism to pass the description of data to and from BNNS functions. The description may include a pointer to the memory location. For example, use the following code when you’re passing immutable weights to a convolution layer: let weights: [Float] = [ ... ] weights.withUnsafeBufferPointer { weightsPtr in let weightsDescriptor = BNNSNDArrayDescriptor(flags: BNNSNDArrayFlags(0), layout: BNNSDataLayoutConvolutionWeightsOIHW, size: (3, 3, 1, 1, 0, 0, 0, 0), stride: (0, 0, 0, 0, 0, 0, 0, 0), data: UnsafeMutableRawPointer(mutating: weightsPtr.baseAddress!), data_type: .float, table_data: nil, table_data_type: .float, data_scale: 1, data_bias: 0) // Create and apply convolution layer. } Setting a stride value of 0 indicates that BNNS calculates stride, without padding, for that axis. For example, the stride for both of the following n-dimensional array descriptors is the same: let inputDescriptor = BNNSNDArrayDescriptor(flags: flags, layout: BNNSDataLayoutRowMajorMatrix, size: (3, 3, 0, 0, 0, 0, 0, 0), stride: (0, 0, 0, 0, 0, 0, 0, 0), data: nil, data_type: .float, table_data: nil, table_data_type: .float, data_scale: 1, data_bias: 0) let inputDescriptor = BNNSNDArrayDescriptor(flags: flags, layout: BNNSDataLayoutRowMajorMatrix, size: (3, 3, 0, 0, 0, 0, 0, 0), stride: (1, 3, 0, 0, 0, 0, 0, 0), data: nil, data_type: .float, table_data: nil, table_data_type: .float, data_scale: 1, data_bias: 0) You don’t need to specify the data when, for example, you’re passing that data directly to BNNSFilterApplyBatch(_:_:_:_:_:_:). The following code creates BNNSNDArrayDescriptor structures for the input and output of a convolution operation. The data property of both descriptors is nil, and the input and output data are passed directly to BNNSFilterApplyBatch(_:_:_:_:_:_:): let input: [Float] = [ ... ] var output: [Float] = [ ... ] let inDescriptor = BNNSNDArrayDescriptor(flags: BNNSNDArrayFlags(0), layout: BNNSDataLayoutImageCHW, size: (6, 6, 1, 0, 0, 0, 0, 0), stride: (0, 0, 0, 0, 0, 0, 0, 0), data: nil, data_type: .float, table_data: nil, table_data_type: .float, data_scale: 1, data_bias: 0) let outDescriptor = BNNSNDArrayDescriptor(flags: BNNSNDArrayFlags(0), layout: BNNSDataLayoutImageCHW, size: (4, 4, 1, 0, 0, 0, 0, 0), stride: (0, 0, 0, 0, 0, 0, 0, 0), data: nil, data_type: .float, table_data: nil, table_data_type: .float, data_scale: 1, data_bias: 0) var parameters = BNNSLayerParametersConvolution(i_desc: inDescriptor, w_desc: weightsDescriptor, o_desc: outDescriptor, bias: biasDescriptor, activation: .identity, x_stride: 1, y_stride: 1, x_dilation_stride: 0, y_dilation_stride: 0, x_padding: 0, y_padding: 0, groups: 1, pad: (0, 0, 0, 0)) // `convolutionLayer` is a `BNNSFilter` created by `BNNSFilterCreateLayerConvolution` using `parameters`. let error = BNNSFilterApplyBatch(convolutionLayer, 1, input, inStride, &output, outStride) ## Topics ### Creating an Array Descriptor - [init(flags:layout:size:stride:data:data_type:table_data:table_data_type:data_scale:data_bias:)](accelerate/bnnsndarraydescriptor/init(flags:layout:size:stride:data:data_type:table_data:table_data_type:data_scale:data_bias:).md) - [init(data:scalarType:shape:)](accelerate/bnnsndarraydescriptor/init(data:scalartype:shape:).md) - [init(data:shape:)](accelerate/bnnsndarraydescriptor/init(data:shape:).md) - [init(dataType:shape:)](accelerate/bnnsndarraydescriptor/init(datatype:shape:).md) - [init()](accelerate/bnnsndarraydescriptor/init().md) ### Specifying the Behavior of an N-Dimensional Array. - [BNNSNDArrayFlags](accelerate/bnnsndarrayflags.md) ### Accessing the Properties of an Array Descriptor - [flags](accelerate/bnnsndarraydescriptor/flags.md) - [layout](accelerate/bnnsndarraydescriptor/layout.md) - [size](accelerate/bnnsndarraydescriptor/size.md) - [stride](accelerate/bnnsndarraydescriptor/stride.md) - [data](accelerate/bnnsndarraydescriptor/data.md) - [data_type](accelerate/bnnsndarraydescriptor/data_type.md) - [table_data](accelerate/bnnsndarraydescriptor/table_data.md) - [table_data_type](accelerate/bnnsndarraydescriptor/table_data_type.md) - [data_scale](accelerate/bnnsndarraydescriptor/data_scale.md) - [data_bias](accelerate/bnnsndarraydescriptor/data_bias.md) - [shape](accelerate/bnnsndarraydescriptor/shape.md) ### Allocating and Deallocating Memory - [allocate(initializingFrom:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(initializingfrom:shape:batchsize:).md) - [allocate(randomUniformUsing:range:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(randomuniformusing:range:shape:batchsize:)-2rorb.md) - [allocate(randomUniformUsing:range:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(randomuniformusing:range:shape:batchsize:)-761hg.md) - [allocate(randomIn:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(randomin:shape:batchsize:)-1697a.md) - [allocate(randomIn:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(randomin:shape:batchsize:)-5a2p2.md) - [allocate(randomIn:using:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(randomin:using:shape:batchsize:)-5kbi8.md) - [allocate(randomIn:using:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(randomin:using:shape:batchsize:)-3w6ig.md) - [allocate(repeating:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(repeating:shape:batchsize:).md) - [allocateUninitialized(scalarType:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocateuninitialized(scalartype:shape:batchsize:).md) - [deallocate()](accelerate/bnnsndarraydescriptor/deallocate().md) ### Generating an Array from an Array Descriptor’s Data - [makeArray(of:batchSize:)](accelerate/bnnsndarraydescriptor/makearray(of:batchsize:).md) ### Initializers - [init(data:scalarType:shape:batchSize:)](accelerate/bnnsndarraydescriptor/init(data:scalartype:shape:batchsize:).md) - [init(data:shape:batchSize:)](accelerate/bnnsndarraydescriptor/init(data:shape:batchsize:).md) ### Instance Properties - [dataSize](accelerate/bnnsndarraydescriptor/datasize.md) ### Type Methods - [allocate(randomNormalUsing:mean:standardDeviation:shape:batchSize:)](accelerate/bnnsndarraydescriptor/allocate(randomnormalusing:mean:standarddeviation:shape:batchsize:).md) ## Relationships ### Conforms To - [BitwiseCopyable](swift/bitwisecopyable.md) ## See Also ### N-dimensional array descriptor essentials - [BNNSLayerData](accelerate/bnnslayerdata.md) - [BNNS.Shape](accelerate/bnns/shape.md) - [BNNSDataLayout](accelerate/bnnsdatalayout.md) - [BNNSDataType](accelerate/bnnsdatatype.md) - [BNNSDataLayoutGetRank(_:)](accelerate/bnnsdatalayoutgetrank(_:).md)