Failing to lower torch model to TOSA

[kslavka-gsi]

Hi,
I am struggling to convert a Torch.nn.Model to TOSA MLIR with the latest torch_mlir builds.
With the torch_mlir snapshot from 27/Jan/2024 the following code works and generates a proper .mlir file.
However, with the latest nightly builds, the .compile is not recogized.
Please suggest how to use the latest torch_mlir to and convert a torch model to TOSA dialect.

import torch
import torch_mlir

class Wrapper(torch.nn.Module):
    def __init__(self) -> None:
        super().__init__()
    
    def forward(self, x, y):
        self.res = torch.add(x,y)
        return self.res

model = Wrapper()
model.eval()
out_tosa_mlir_path = "./vecadd_tosa.mlir"
x = torch.randn(64 * 1024)
y = torch.randn(64 * 1024)
alpha = 2

module = torch_mlir.compile(model, [x,y],output_type=torch_mlir.OutputType.TOSA, use_tracing=True)
with open(out_tosa_mlir_path, "w", encoding="utf-8") as outf:
    outf.write(str(module))