Add pre-processing in PredictorTRT pipeline

[zhiqwang]

We introduce a new PredictorTRT inference pipeline to call the exported TensorRT engine.

import torch
from yolort.runtime import PredictorTRT

# Load the exported TensorRT engine
engine_path = 'yolov5n6.engine'
device = torch.device('cuda')
y_runtime = PredictorTRT(engine_path, device=device)

# Perform inference on an image file
predictions = y_runtime.predict('bus.jpg')

[codecov]

Codecov Report

Merging #309 (dfdd861) into main (720cd32) will not change coverage.
The diff coverage is n/a.

❗ Current head dfdd861 differs from pull request most recent head 320ec86. Consider uploading reports for the commit 320ec86 to get more accurate results

@@           Coverage Diff           @@
##             main     #309   +/-   ##
=======================================
  Coverage   94.93%   94.93%           
=======================================
  Files          11       11           
  Lines         731      731           
=======================================
  Hits          694      694           
  Misses         37       37           

Flag	Coverage Δ
unittests	94.93% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 720cd32...320ec86. Read the comment docs.

[CLAassistant]

All committers have signed the CLA.