🐛 [Bug] Error encountered when compiling Waveglow model

[gs-olive]

Bug Description

When compiling the Waveglow (Github) spectrogram-to-speech model with Torch-TRT, the following error is encountered:

RuntimeError: 
Schema not found for node. File a bug report.
Node: %25626 : int = aten::floor_divide(%22918, %25625)

Input types:int, int
candidates were:
  aten::floor_divide(Tensor self, Tensor other) -> Tensor
  aten::floor_divide.Scalar(Tensor self, Scalar other) -> Tensor
  aten::floor_divide.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)

To Reproduce

Steps to reproduce the behavior:

1. Run torch_tensorrt.compile with Waveglow model as input, using fp32 precision.
2. Choose fixed input sizes of (1, 80, 256) and (1, 8, 8192) and enable truncate_long_and_double with 12 GB workspace.

Expected behavior

Model should successfully compile to Torch-TRT. Specifically, this invalid aten::floor_divide schema should not arise in the TorchScript IR.

Environment

• Torch-TensorRT Version: 1.3.0a0+e3b99294
• PyTorch Version: 1.13.0.dev20220921+cu116
• CPU Architecture: Intel Xeon CPU
• OS: Ubuntu 20.04
• How you installed PyTorch: pip
• Build command you used: python setup.py develop
• Are you using local sources or building from archives: local
• Python version: 3.8.13
• CUDA version: 11.6