MetalPerformanceShaders Namespace

A image kernel that combines two textures into one texture result.

A MPSCnnKernel that has binary weights and convolves its input.

A MPSCnnConvolutionNode that represents a binary convolution kernel.

A MPSCnnBinaryConvolution that is a fully-connected convolution layer that uses binary weights.

A MPSCnnBinaryConvolutionNode that represents a fully-connected convolution layer that uses binary weights.

A MPSKernel that has binary weights.

A MPSCnnKernel that convolves its inputs, producing a feature map for each of its constituent filters.

Base class for classes that provide weights and bias terms to convolution filters.

Extension methods to the IMPSCnnConvolutionDataSource interface to support all the methods from the MPSCnnConvolutionDataSource protocol.

Describes a convolution kernel.

Subclass of MPSNNFilterNode that is the base class for convolution representation nodes.

A MPSCnnKernel that transposes its input.

A MPSCnnConvolutionNode that represents a transpose kernel.

A MPSCnnKernel that normalizes across feature channels.

A MPSCnnNormalizationNode that represents a cross-channel normalization kernel.

A MPSCnnConvolutionDescriptor that describes depthwise convolution.

A dilated max MPSCnnPooling filter.

A MPSNNFilterNode that represents a dilated max pooling filter.

A fully connected layer.

A MPSCnnConvolutionNode that represents a fully connected layer.

The base class for layers in a convolutional neural network.

A local contrast MPSCnnKernel.

A MPSCnnNormalizationNode representing a local contrast normalization kernel.

The logarithmic softmax loss function.

A MPSNNFilterNode that represents the logarithmic softmax loss function.

The base class for activation functions.

The absolute-value activation function.

A MPSCnnNeuronNode that represents the absolute-value activation function.

The exponential linear unit activation function.

A MPSCnnNeuronNode that represents the exponential linear unit activation function.

The hard sigmoid activation function.

A MPSCnnNeuronNode that represents the hard sigmoid activation function.

The linear activation function.

A MPSCnnNeuronNode that represents the linear activation function.

The base class for representations of activation functions.

The parametric rectified linear unit activation function.

A MPSCnnNeuronNode that represents the parametric rectified linear unit activation function.

The rectified linear unit activation function.

The ReLUN activation function.

A MPSCnnNeuronNode that represents the ReLUN activation function.

A MPSCnnNeuronNode that represents the rectified linear unit activation function.

The sigmoid activation function.

A MPSCnnNeuronNode that represents the sigmoid activation function.

The softplus activation function.

A MPSCnnNeuronNode that represents the softplus activation function.

The softsign activation function.

A MPSCnnNeuronNode that represents the softsign activation function.

The hyperbolic tangent activation function.

A T:MetalPerformanceShaders.MPSCNNNeuronNode that represents a tanh activation function.

Abstract base class for normalization MPSNNFilterNode subclasses.

A subclass of MPSCnnKernel that sub-samples its input.

An average pooling filter.

A MPSNNFilterNode that returns the average value of its input region.

An L2-norm pooling filter.

Representation of an L2-norm pooling filter.

A MPSCnnPooling subclass that performs max-pooling.

A MPSCnnPoolingNode that represents a max-pooling kernel in a convolutional neural net.

A MPSNNFilterNode that represents a pooling kernel in a convolutional neural net.

A MPSCnnKernel commonly used in categorization tasks.

A MPSNNFilterNode that represents a softmax filter.

A MPSCnnKernel that performs spatial normalization.

A MPSCnnNormalizationNode that represents a spatial normalization kernel.

A MPSCnnConvolutionDescriptor that describes subpixel upsampling and reshaping.

A MPSCnnKernel that upsamples an image.

A MPSCnnUpsampling filter that performs bilinear spatial upsampling.

A MPSNNFilterNode that performs bilinear spatial upsampling.

A MPSCnnUpsampling filter that performs nearest spatial upsampling.

A MPSNNFilterNode that performs nearest spatial upsampling.

Describes a gated recurrent unit layer in a neural net.

An image that may contain more than 4 channels. (For example, an image in a layer of a convolutional neural network.)

A MPSImageArithmetic kernel that performs element-wise addition of two images.

Filter that finds the maximum pixel value in a window around each pixel in the source image.(Individual channels are processed separately.)

Filter that finds the minimum pixel value in a window around each pixel in the source image.(Individual channels are processed separately.)

Base class for MPSBinaryImageKernel classes that perform arithmetic on images.

A MPSImageScale subclass that uses bilinear sampling to scale the image.

Filter that blurs by transforming each pixel of the source image to the average of itself and its neighbors.

An image that can convert its color space, alpha, or pixel format.

Filter that represents a convolution.

A MPSKernel that copies image data into a MPSMatrix object.

Contains the attributes for a new or existing MPSImage.

Finds the maximum value in a region, offset by a corresponding value in a mask.

A MPSImageArithmetic kernel that performs element-wise division of two images.

Finds the mininum value in a region, offset by a corresponding value in a mask.

Filter that applies a fast Gaussian blur to an image.

Represents a Gaussian image pyramid.

Calculates a histogram of image data.

A MPSUnaryImageKernel that equalizes the histogram of an image.

Transforms an image so that its histogram matches a desired histogram.

Filter that sums the values of pixels in a region.

Filter that sums the squared values of pixels in a region.

Filter that performs a scaling operation with Lanczos resampling.

An optimized Laplacian filter.

Filter that finds the median value of each channel for pixels in the region around each source image pixel.

A MPSImageArithmetic kernel that performs element-wise multiplication of two images.

Base class for image pyramids.

A MPSUnaryImageKernel that can resize and change aspect ratio of an image.

Filter that detects edges by using a Sobel filter.

A MPSUnaryImageKernel that calculates the mean of pixel values for a region.

A MPSUnaryImageKernel that calculates the mean and variance of pixel values for a region.

A MPSUnaryImageKernel that calculates the minimum and maximum pixel values for a region.

A MPSImageArithmetic kernel that performs element-wise subtraction of two images.

Filter that blurs an image with a tent function.

Filter that changes all pixels above a threshold luminance to a specified maximum single-channel value, and completely darkens the rest.

Filter that changes all pixels above a threshold luminance to 0, and brightens the rest to a specified maximum single-channel value.

Filter that darkens all pixels dimmer than or equal in brightness to a threshold luminance to 0, and leaves the rest unchanged.

Filter that leaves all pixels dimmer than or equal in brightness to a threshold luminance unchangedt, and darkens the rest to 0.

Filter that clamps brightness values to a threshold value.

Filter that transposes an image.

Base class that represents the kernel of a shader.

Describes a Long-Short Term Memory layer in neural net.

A matrix that represents the kernel of a linear transformation.

A kernel that operates on two matrices to create a new matrix.

Performs multiple matrix copy operations.

Describes multiple matrix copy operations.

A MPSMatrixUnaryKernel that computes the Cholesky factorization.

A MPSMatrixUnaryKernel that computes LU factorization using partial pivoting.

Describes the size, data type, and stride of a row-major matrix.

Represents a weighted matrix multiplication operation, followed by a weighted addition operation.

A MPSMatrixBinaryKernel that solves a linear system of equations via Cholesky factorization.

A MPSMatrixBinaryKernel that solves a linear system of equations via a triangular coefficient matrix.

A kernel that performs a mapping from one matrix to another.

Performs matrix multiplication.

Adds the results of two kernels.

A MPSNNFilterNode that performs bilinear sampling.

Abstract base class of arithmetic nodes.

Concatenates the results of two kernels.

Predefined common padding policies.

Divides the results of two kernels.

A placeholder node in a neural network graph for an image filtering stage.

An optimized neural network graph.

A placeholder node for an image in a neural network graph.

A MPSNNFilterNode that performs Lanczos resampling.

Multiplies the results of two kernels.

Abstract neural network graph node for image resampling.

A state object in a neural network graph.

Subtracts the results of two kernels.

A structural description of a recurrent neural net layer.

A recurrent neural net layer specifically for image data.

A recurrent neural net layer.

The image containing the state in an image-baed recurrent neural net as it iterates through its sequence.

The matrix containing the state as a recurrent neural net iterates through its sequence.

Describes the internal gate in a recurrent neural net.

Temporary storage used by convolutional neural nets.

Discardable image data.

A matrix allocated in GPU private memory.

A one-dimensional array allocated in GPU private memory.

Represents a shader transformation produces one texture from another.

A one-dimensional array.

Describes the length and data type of a MPSVector.

Specifies the range of histogram data in a histogram, the number of entries, and whether to encode the alpha channel.

Options for the discovery of keypoints in an image.

Options for the reading and writing of feature channels in an image.

Describes a copy operation that supports offsets.

A coordinate that represents an offset.

A coordinate that represents the origin of a coordinate system.

Structure that represents a region as an origin and a size.

A transformation for use with a Lanczos kernel.

A structure that represents a width, height, and depth.

Interface representing the required methods (if any) of the protocol MPSCnnConvolutionDataSource.

Interface describing a Metal Performance Shaders-specific identifier.

Interface defining a factory that generates a MPSImage from a T:MetalPerformanceShaders.MTLCommandBuffer, a MPSImageDescriptor, and a MPSKernel.

Interface defining methods relating to when image size is stored elsewhere in the graph.

Interface defining image resampling methods.

Interface describing how kernels should pad their inputs.

Enumerates values that indicate if and what kind of color premultiplication will be applied to color values.

Flagging enumeration for options available to binary convolution kernels.

Enumerates the operation used in a binary convolution.

Developers should not use this deprecated class.

Enumerates the available activation functions of a neuron.

Enumerates whether a data buffer is row- or column-major

Enumerates values that specify floating point data types.

Enumerates shader behavior at the edges of regions and images.

Enumerates image channel descriptions.

Enumerates ORable kernel options that improve performance in certain cases.

Enumerates the result forms of a matrix decomposition.

Options for how a neural network graph will pad results.

Enumerates an image's underlying texture's purgeable state.

Enumerates how input matrices or images should be combined in a recurrent neural net.

Enumerates the propagation direction in a layer in a recurrent neural net.

Commands to copy a source texture to a new location. Used for out-of-place filters.