---
title: "init(type:kernelSizes:inputFeatureChannelCount:outputFeatureChannelCount:groupCount:strides:dilationRates:paddingPolicy:)"
framework: ML Compute
role: symbol
role_heading: Initializer
platforms: [iOS 14.0+, iPadOS 14.0+, Mac Catalyst 14.0+, macOS 11.0+, tvOS 14.0+]
path: "mlcompute/mlcconvolutiondescriptor/init(type:kernelsizes:inputfeaturechannelcount:outputfeaturechannelcount:groupcount:strides:dilationrates:paddingpolicy:)"
---

# init(type:kernelSizes:inputFeatureChannelCount:outputFeatureChannelCount:groupCount:strides:dilationRates:paddingPolicy:)

Creates a descriptor for the type, sizes, number of feature channels, group count, strides, dilation rates, and padding policy you specify.

## Declaration

```swift
convenience init(type: MLCConvolutionType = .standard, kernelSizes: (height: Int, width: Int), inputFeatureChannelCount: Int, outputFeatureChannelCount: Int, groupCount: Int = 1, strides: (y: Int, x: Int) = (1, 1), dilationRates: (y: Int, x: Int) = (1, 1), paddingPolicy: MLCPaddingPolicy = .same)
```

## Parameters

- `type`: The convolution type.
- `kernelSizes`: A tuple that contains the kernel sizes for y and x.
- `inputFeatureChannelCount`: The number of feature channels in the input tensor.
- `outputFeatureChannelCount`: The number of feature channels in the output tensor.
- `groupCount`: The number of groups.
- `strides`: A tuple that contains the kernel strides for y and x.
- `dilationRates`: A tuple that contains the dilation rates for y and x.
- `paddingPolicy`: The padding policy.

## See Also

### Creating Convolution Descriptors

- [init(transposeWithKernelWidth:kernelHeight:inputFeatureChannelCount:outputFeatureChannelCount:)](mlcompute/mlcconvolutiondescriptor/init(transposewithkernelwidth:kernelheight:inputfeaturechannelcount:outputfeaturechannelcount:).md)
- [init(depthwiseWithKernelWidth:kernelHeight:inputFeatureChannelCount:channelMultiplier:)](mlcompute/mlcconvolutiondescriptor/init(depthwisewithkernelwidth:kernelheight:inputfeaturechannelcount:channelmultiplier:).md)
- [MLCConvolutionType](mlcompute/mlcconvolutiontype.md)