00000220  6f 20 52 6f 6f 74 30 1e  17 0d 31 36 30 31 30 31  |o Root0...160101|
00000230  30 30 30 30 30 30 5a 17  0d 32 35 30 31 30 31 30  |000000Z..2501010|
00000240  30 30 30 30 30 5a 30 1a  31 0b 30 09 06 03 55 04  |00000Z0.1.0...U.|
00000250  0a 13 02 47 6f 31 0b 30  09 06 03 55 04 03 13 02  |...Go1.0...U....|
00000260  47 6f 30 81 9f 30 0d 06  09 2a 86 48 86 f7 0d 01  |Go0..0...*.H....|

000000c0  6f 20 52 6f 6f 74 30 1e  17 0d 31 36 30 31 30 31  |o Root0...160101|
000000d0  30 30 30 30 30 30 5a 17  0d 32 35 30 31 30 31 30  |000000Z..2501010|
000000e0  30 30 30 30 30 5a 30 1a  31 0b 30 09 06 03 55 04  |00000Z0.1.0...U.|
000000f0  0a 13 02 47 6f 31 0b 30  09 06 03 55 04 03 13 02  |...Go1.0...U....|
00000100  47 6f 30 81 9f 30 0d 06  09 2a 86 48 86 f7 0d 01  |Go0..0...*.H....|

  %q2 = "quantfork.qcast"(%cst) {volatile} : (tensor<3x2xf32>) -> tensor<3x2x!quant.uniform<i8<-127:127>:f32, 5.000000e-03>>
  %dot = stablehlo.dot %q1, %q2 : (tensor<1x3x!quant.uniform<i8:f32, 6.000000e-03:-128>>, tensor<3x2x!quant.uniform<i8<-127:127>:f32, 5.000000e-03>>) -> tensor<1x2x!quant.uniform<i8:f32, 1.000000e-03:-3>>

00000220  fa 10 e2 59 ec 47 2d 7c  ac da 4e 97 0e 15 a0 6f  |...Y.G-|..N....o|
00000230  d0 02 42 01 4d fc be 67  13 9c 2d 05 0e bd 3f a3  |..B.M..g..-...?.|
00000240  8c 25 c1 33 13 83 0d 94  06 bb d4 37 7a f6 ec 7a  |.%.3.......7z..z|
00000250  c9 86 2e dd d7 11 69 7f  85 7c 56 de fb 31 78 2b  |......i..|V..1x+|
00000260  e4 c7 78 0d ae cb be 9e  4e 36 24 31 7b 6a 0f 39  |..x.....N6$1{j.9|
00000270  95 12 07 8f 2a 16 03 01  00 b4 0c 00 00 b0 03 00  |....*...........|

  %0 = stablehlo.convolution(%arg0, %arg1) dim_numbers = [b, f, 0, 1]x[o, i, 0, 1]->[b, f, 0, 1], window = {pad = [[1, 1], [1, 1]]} {batch_group_count = 1 : i64, feature_group_count = 1 : i64} : (tensor<1x4x3x3x!quant.uniform<i8:f32, 1.000000e+0:-100>>, tensor<2x4x3x3x!quant.uniform<i8:f32:0, {2.000000e+2,...

// CHECK: %[[QUANTIZE:.*]] = "quantfork.qcast"(%[[CONSTANT0]]) : (tensor<3x3x3x16xf32>) -> tensor<3x3x3x16x!quant.uniform<i8:f32:3,
// CHECK-SAME: {1.000000e+00:-128,1.000000e+00:-128,1.000000e+00:-128,1.000000e+00:-128,1.000000e+00:-128,1.000000e+00:-128,1.000000e+00:-128,1.000000e+00:-128,1.000000e+00:-128,

  // CHECK-DAG: %[[FILTER:.+]] = mhlo.constant dense<{{\[\[\[\[}}1.000000e+00, 2.000000e+00], [3.000000e+00, 4.000000e+00], [5.000000e+00, 6.000000e+00]]]]> : tensor<1x1x3x2xf32>
  // CHECK-DAG: %[[CST_0:.+]] = mhlo.constant dense<[1.000000e-01, 2.000000e-01]> : tensor<2xf32>
  // CHECK-DAG: %[[CST_1:.+]] = mhlo.constant dense<[3.000000e-01, 4.000000e-01]> : tensor<2xf32>

// CHECK: %[[QUANTIZE:.*]] = "quantfork.qcast"(%[[CONSTANT0]]) : (tensor<3x3x3x16xf32>) -> tensor<3x3x3x16x!quant.uniform<i4:f32:3,
// CHECK-SAME: {1.000000e+00:-8,1.000000e+00:-8,1.000000e+00:-8,1.000000e+00:-8,1.000000e+00:-8,1.000000e+00:-8,1.000000e+00:-8,1.000000e+00:-8,1.000000e+00:-8,

    %0 = "tf.FakeQuantWithMinMaxArgs"(%arg0) {device = "", max = 2.000000e-01 : f32, min = -1.000000e-01 : f32, narrow_range = false, num_bits = 8 : i64} : (tensor<1x3x4x3xf32>) -> tensor<1x3x4x3xf32>

  %0 = stablehlo.constant dense<[[0.000000e+0, 1.000000e+0, 2.000000e+0], [3.000000e+0, 4.000000e+0, 5.000000e+0]]> : tensor<2x3xf32>
  %1 = stablehlo.transpose %0, dims = [1, 0] : (tensor<2x3xf32>) -> tensor<3x2xf32>
  return %1 : tensor<3x2xf32>
}
// CHECK-DAG: %[[CONST_0:.+]] = stablehlo.constant dense<{{\[\[}}0.000000e+00, 3.000000e+00], [1.000000e+00, 4.000000e+00], [2.000000e+00, 5.000000e+00]]> : tensor<3x2xf32>
// CHECK-NOT: transpose