[circle-mlir] Introduce unit test model for Transpose

circle-mlir/models/unit/Transpose_F32_R4/__init__.py

@@ -0,0 +1,21 @@
+import torch
+
+
+# Generate Transpose operator with Float32, Rank-4
+class net_Transpose(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+
+    def forward(self, input):
+        # NOTE two transpose will produce NCHW to NHWC single transpose
+        x = torch.transpose(input, 1, 3)
+        return torch.transpose(x, 1, 2)
+
+    def onnx_opset_version(self):
+        # TODO set to appropriate value
+        return 13
+
+
+_model_ = net_Transpose()
+
+_inputs_ = torch.randn(1, 3, 6, 4)