I need to reshape a tensor A to shape S that is not available at the compile time. I would like to compute the shape S inside the graph and then reshape the tensor A to the shape S. Is this even possible with MPS?

I would like to generate and run ML program inside an app. I got familiar with the coremltools and MIL format, however I can't seem to find any resources on how to generate mlmodel/mlpackage files using Swift on the device. Is there any Swift equivalent of coremltools? Or is there a way to translate MIL description of a ML program into instance of a MLModel? Or something similar.

Hello, printing the dataType property some tensor, results in a non-disclosing "MPSDataType" string. Is this intended? If so, how do I get string representation of type of the tensor please? let graph = MPSGraph() print(graph.constant(23.0, dataType: .float32).dataType) // prints "MPSDataType", and not "f32" or something useful

Hello, I am trying to implement numpy.arrange using MPSGraph primitives. Along the way I got to this code, which I would expect is correct, however the program crashes with information about type mismatch. What am I doing wrong? Code let G = MPSGraph() let length = G.constant(9, shape: [1], dataType: .int32) let base = G.constant(1, shape: [1], dataType: .int32) let template = G.broadcast(base, shapeTensor: length, name: nil) let r = G.for(     numberOfIterations: length,     initialBodyArguments: [template], // just return the argument     body: { i, args in [args[0]] },     name: nil ) Error -:22:11: error: 'scf.for' op types mismatch between 0th iter operand and defined value -:22:11: note: see current operation: %19 = "scf.for"(%16, %17, %18, %12) ( { ^bb0(%arg1: index, %arg2: tensor<*xi32>):  // no predecessors   %21 = "tensor.from_elements"(%arg1) : (index) -> tensor<1xindex>   %22 = "mps.cast"(%21) {resultElementType = i32} : (tensor<1xindex>) -> tensor<1xi32>   "scf.yield"(%arg2) : (tensor<*xi32>) -> () }) : (index, index, index, tensor<9xi32>) -> tensor<*xi32> /Library/Caches/com.apple.xbs/Sources/MetalPerformanceShadersGraph_Sim/MetalPerformanceShadersGraph-2.0.22/mpsgraph/MetalPerformanceShadersGraph/Core/Files/MPSGraphExecutable.mm:1052: failed assertion `Error: MLIR pass manager failed' P.S. Is there a way I can return differently-shaped tensor from each iteration? In other words: body: { i, args in [G.concatTensors([args[0], i], dimension: 0, name: nil)] }

Is it possible to have object texture that is different for each eye? Or maybe having a eye index information in the shader? With rendering slightly different colors for each eye, you can achieve extremely satisfying visuals (iridescence/holographic-like) or increase perceptual contrast where it is needed.