Any chance to quantize a torchscript model using pt2e?

quantization
1

I am exploring pt2e quantization and want to quantize a torchscript model using pt2e, but I failed to convert a torchscript to torch.export.

I wonder if there is any chance to quantize a torchscript model using pt2e? Thanks.

import torch

m = torch.nn.Conv2d(3, 3, 3)
m = torch.jit.script(m)
inputs = (torch.rand(1, 3, 32, 32), )
m(*inputs)

import torch.export._trace
torch.export._trace._export(m, inputs, pre_dispatch=True)

Error message:

 gm_torch_level, _ = torch._dynamo.export(
  File ".../python3.8/site-packages/torch/_dynamo/eval_frame.py", line 1356, in inner
    assert (
AssertionError: Failed to produce a graph during tracing as no tensor operations were found.