tensorflow - 使用 Metal 性能着色器解卷积

Question

原来没有这样的操作 deconvolution在 MPS . tensorflow中最接近的类似物是 conv2d_transpose .

是否可以在MPS之间排序插件自定义操作默认操作？

Answer 1

您可以编写自己的 Metal 计算内核并在 MPS 操作之间执行这些内核。

例如:

let commandBuffer = commandQueue.makeCommandBuffer()

. . .

// Do something with an MPSCNN layer:
layer1.encode(commandBuffer: commandBuffer, sourceImage: img1, destinationImage: img2)

// Perform your own compute kernel:
let encoder = commandBuffer.makeComputeCommandEncoder()
encoder.setComputePipelineState(yourOwnComputePipeline)
encoder.setTexture(img2.texture, at: 0)
encoder.setTexture(img3.texture, at: 1)
let threadGroupSize = MTLSizeMake(. . .)
let threadGroups = MTLSizeMake(img2.texture.width / threadGroupSize.width,
                               img2.texture.height / threadGroupSize.height, 1)
encoder.dispatchThreadgroups(threadGroups, threadsPerThreadgroup: threadGroupSize)
encoder.endEncoding()

// Do something with another MPSCNN layer:
layer2.encode(commandBuffer: commandBuffer, sourceImage: img3, destinationImage: img4)

. . .

commandBuffer.commit()

您必须使用 Metal Shading Language 编写自己的计算内核并将其加载到 yourOwnComputePipeline 中。目的。然后您可以随时将其编码到当前命令缓冲区中。