Edit this page

Converting a ONNX* Model

Introduction to ONNX

ONNX* is a representation format for deep learning models. ONNX allows AI developers easily transfer models between different frameworks that helps to choose the best combination for them. Today, PyTorch*, Caffe2*, Apache MXNet*, Microsoft Cognitive Toolkit* and other tools are developing ONNX support.

Supported Public ONNX Topologies

Model Name	Path to Public Models master branch
bert_large	model archive
bvlc_alexnet	model archive
bvlc_googlenet	model archive
bvlc_reference_caffenet	model archive
bvlc_reference_rcnn_ilsvrc13	model archive
inception_v1	model archive
inception_v2	model archive
resnet50	model archive
squeezenet	model archive
densenet121	model archive
emotion_ferplus	model archive
mnist	model archive
shufflenet	model archive
VGG19	model archive
zfnet512	model archive
GPT-2	model archive
YOLOv3	model archive

Listed models are built with the operation set version 8 except the GPT-2 model (which uses version 10). Models that are upgraded to higher operation set versions may not be supported.

Supported PaddlePaddle* Models via ONNX Conversion

Starting from the R5 release, the OpenVINO™ toolkit officially supports public PaddlePaddle* models via ONNX conversion. The list of supported topologies downloadable from PaddleHub is presented below:

Model Name	Command to download the model from PaddleHub
MobileNetV2	hub install mobilenet_v2_imagenet==1.0.1
ResNet18	hub install resnet_v2_18_imagenet==1.0.0
ResNet34	hub install resnet_v2_34_imagenet==1.0.0
ResNet50	hub install resnet_v2_50_imagenet==1.0.1
ResNet101	hub install resnet_v2_101_imagenet==1.0.1
ResNet152	hub install resnet_v2_152_imagenet==1.0.1

Note

To convert a model downloaded from PaddleHub use paddle2onnx converter.

The list of supported topologies from the models v1.5 package:

• MobileNetV1

• MobileNetV2

• ResNet

• ResNet_vc

• ResNet_vd

• ResNeXt

• ResNeXt_vd

Note

To convert these topologies one should first serialize the model by calling paddle.fluid.io.save_inference_model

(description) command and after that use paddle2onnx converter.

Convert an ONNX* Model

The Model Optimizer process assumes you have an ONNX model that was directly downloaded from a public repository or converted from any framework that supports exporting to the ONNX format.

To convert an ONNX* model, run Model Optimizer with the path to the input model .nnet file and an output directory where you have write permissions:

Package, Docker, open-source installation

cd <INSTALL_DIR>/deployment_tools/model_optimizer/
python3 mo.py --input_model <INPUT_MODEL>.onnx --output_dir <OUTPUT_MODEL_DIR>

pip installation

mo --input_model <INPUT_MODEL>.onnx --output_dir <OUTPUT_MODEL_DIR>

There are no ONNX* specific parameters, so only framework-agnostic parameters are available to convert your model. For details, see see the General Conversion Parameters section on the Converting a Model to Intermediate Representation (IR) page.

Supported ONNX* Layers

Refer to Supported Framework Layers for the list of supported standard layers.

Prev Next