Scripted inception_v3 has wrong output format

🐛 Describe the bug

Scripted inception_v3 model has wrong output format - Instead of Tensor it returns InceptionOutputs.
I checked other models - resnet50, mobilenet_v2 - After scripting they return tensor.
Inference tools such as torch_tensorrt expects models to return Tensor or [Tensor].

To reproduce the issue with inception_v3:

import torch
import torchvision.models as models

model = models.inception_v3(pretrained=True).eval()
x = torch.rand(1, 3, 299, 299)
y=model(x)
smodel = torch.jit.script(model, x).eval()
y=smodel(x)

print(type(y))

It prints <class 'torch._jit_internal.InceptionOutputs'>.

Expected output: <class 'torch.Tensor'>

What should be fixed in inception.py code to make it output Tensor after applying torch.jit.script?

Versions

PyTorch version: 1.12.0a0+bd13bc6
Is debug build: False
CUDA used to build PyTorch: 11.6
ROCM used to build PyTorch: N/A

OS: Ubuntu 20.04.4 LTS (x86_64)
GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
Clang version: Could not collect
CMake version: version 3.22.3
Libc version: glibc-2.31

Python version: 3.8.13 | packaged by conda-forge | (default, Mar 25 2022, 06:04:10)  [GCC 10.3.0] (64-bit runtime)
Python platform: Linux-5.4.0-1075-aws-x86_64-with-glibc2.10
Is CUDA available: True
CUDA runtime version: 11.6.124
GPU models and configuration: GPU 0: Tesla T4
Nvidia driver version: 515.43.04
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.4.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.4.0
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

Versions of relevant libraries:
[pip3] numpy==1.22.3
[pip3] pytorch-quantization==2.1.2
[pip3] torch==1.12.0a0+bd13bc6
[pip3] torch-tensorrt==1.1.0a0
[pip3] torchtext==0.13.0a0
[pip3] torchvision==0.13.0a0
[conda] magma-cuda110             2.5.2                         5    local
[conda] mkl                       2019.5                      281    conda-forge
[conda] mkl-include               2019.5                      281    conda-forge
[conda] numpy                     1.22.3           py38h1d589f8_2    conda-forge
[conda] pytorch-quantization      2.1.2                    pypi_0    pypi
[conda] torch                     1.12.0a0+bd13bc6          pypi_0    pypi
[conda] torch-tensorrt            1.1.0a0                  pypi_0    pypi
[conda] torchtext                 0.13.0a0                 pypi_0    pypi
[conda] torchvision               0.13.0a0                 pypi_0    pypi

Start in torchvision/models/inception.py at the inception_v3 entry point and run the reproduction script from the issue with a scripted model. Compare the scripted result with the eager model and the resnet50 or mobilenet_v2 examples. Done means scripted inception_v3 returns a Tensor rather than InceptionOutputs.