Inconsistent inference results with Conv3d operator

[Thrsu]

Description:

I'm experiencing a discrepancy between the inference results of my PyTorch model and the TensorRT model obtained by converting it using the torch2trt tool.

Reproduce

This problem can be reproduced by the following script:

from torch2trt import torch2trt
import torch
from torch.nn import Module

model = torch.nn.Conv3d(2,3,3,2,(1, 2, 3),1,1,True,'circular',).cuda()
input_data = torch.randn([2, 2, 4, 4, 4], dtype=torch.float32).cuda()
model_trt = torch2trt(model, [input_data])
y = model(input_data)
y_trt = model_trt(input_data)

# check the output against PyTorch
print(torch.max(torch.abs(y - y_trt)))
 

The output is:

tensor(1.6616, device='cuda:0', grad_fn=<MaxBackward1>)

Environment

• torch: 1.11.0
• torch2trt: 0.4.0
• tensorrt: 8.6.1.6