Skip to content

Installation and Usage of the Paddle2ONNX Plugin

The Paddle2ONNX plugin for PaddleX provides the ability to convert PaddlePaddle static models to ONNX format models, leveraging the underlying Paddle2ONNX.

1. Installation

# Windows users need to use the following command to install the dev version of paddlepaddle
# python -m pip install --pre paddlepaddle -i https://www.paddlepaddle.org.cn/packages/nightly/cpu/

# Install Paddle2ONNX plugin
paddlex --install paddle2onnx

2. Usage

2.1 Parameter Introduction

Parameter Type Description
paddle_model_dir str Directory containing the Paddle model.
onnx_model_dir str Output directory for the ONNX model, which can be the same as the Paddle model directory. Defaults to onnx.
opset_version int The ONNX opset version to use. If a lower-version opset cannot complete the conversion, a higher-version opset will be automatically selected for the conversion. Defaults to 7.

2.2 Usage Method

Usage:

paddlex \
    --paddle2onnx \  # Use the paddle2onnx function
    --paddle_model_dir /your/paddle_model/dir \  # Specify the directory where the Paddle model is located
    --onnx_model_dir /your/onnx_model/output/dir \  # Specify the output directory for the converted ONNX model
    --opset_version 7  # Specify the ONNX opset version to use

Taking the ResNet18 model from the image_classification module as an example:

paddlex \
    --paddle2onnx \
    --paddle_model_dir ./ResNet18 \
    --onnx_model_dir ./ResNet18 \