Article ID: 000088869 Content Type: Troubleshooting Last Reviewed: 09/08/2022

How Can I Improve the Inferencing Performance of the YOLOv4 Model?

BUILT IN - ARTICLE INTRO SECOND COMPONENT
Summary

Use the Post-Training Optimization Tool (POT) to accelerate the inference of deep learning models.

Description
  • Trained a YOLOv4 model with non-square images using PyTorch.
  • Converted the weights to ONNX file and then to Intermediate Representation (IR).
  • Unable to determine how to achieve better inferencing performance.
Resolution

The Post-Training Optimization Tool (POT) is designed to accelerate the inference of deep learning models by applying special methods without model retraining or fine-tuning.

Additional information

Refer to Training on non-square images and Rectangular Inference for how to implement non-square trained images on YOLO model.

Related Products

This article applies to 2 products