MLIR support

[victorygogogo]

how to support mlir file about netron?　

[lutzroeder]

@victorygogogo can you provide some context. Are you trying to create a pull request to to add MLIR support? Can you share a sample model file, how it was created and how it is used.

[victorygogogo]

mlir file content as flow:
module {
func.func @test_onnx_conv_simple_pattern(%arg0: tensor<5x3x32x32xf32>, %arg1: tensor<?x3x2x2xf32>) -> tensor<5x?x31x31xf32> {
%0 = "onnx.NoValue"() {value} : () -> none
%1 = "onnx.Conv"(%arg0, %arg1, %0) {auto_pad = "NOTSET", kernel_shape = [2, 2], pads = [0, 0, 0, 0]} : (tensor<5x3x32x32xf32>, tensor<?x3x2x2xf32>, none) -> tensor<5x?x31x31xf32>
return %1 : tensor<5x?x31x31xf32>
}
}
I am trying to find some tool to support mlir file to visualize as netron, if netron support is best.

[lutzroeder]

@victorygogogo can you implement MLIR support and submit a pull request. Both a JavaScript parser and code to translate into objects for visualization can be added to mlir.js.

[lutzroeder]

Duplicate of #1044