Deep Learning Import and Export

将网络从和导出网络导入到外部深度学习平台

导入网络和网络体系结构从Tensorflow™-Keras，Tensorflow 2，ONNX™（开放神经网络交换）模型格式和CAFFE。您还可以将经过训练的深度学习Toolbox™网络导出到ONNX模型格式。有关更多信息，请参阅预处理的深神经网络.

You must have support packages to run the import and export functions in Deep Learning Toolbox. If the support package is not installed, each function provides a download link to the corresponding support package in the Add-On Explorer. A recommended practice is to download the support package to the default location for the version of MATLAB^®you are running. You can also directly download the support packages from the following links.

TheimportONNXNetwork,ExpientOnnXlayers,importONNXFunction，和exportONNXNetwork功能需要Deep Learning Toolbox Converter for ONNX Model Format. To download the support package, go to//www.tatmou.com/matlabcentral/fileexchange/67296-deep-learning-toolbox-converter-for-onnx-model-format.
TheImportTensorFlownetwork,ImportTensorFlowlayers,ImportKerasnetwork，和ImportKeraslayersrequire深度学习工具箱转换器用于张量流模型. To download the support package, go to//www.tatmou.com/matlabcentral/fileexchange/64649-deep-learning-toolbox-converter-for-tensorflow-models.
TheimportCaffeNetwork和importCaffeLayers功能需要Deep Learning Toolbox Importer for Caffe Models. To download the support package, go to//www.tatmou.com/matlabcentral/fileexchange/61735-deep-learning-toolbox-importer-for-caffe-models.

Functions

expand all

Network Import

TensorFlowImport

`ImportTensorFlownetwork`	Import pretrainedTensorFlow网络
`ImportTensorFlowlayers`	Import layers fromTensorFlow网络
`ImportKerasnetwork`	Import pretrained Keras network and weights
`ImportKeraslayers`	Import layers from Keras network

caffe进口

`importCaffeNetwork`	Import pretrained convolutional neural network models from Caffe
`importCaffeLayers`	Import convolutional neural network layers from Caffe

onnxImport

`importONNXNetwork`	Import pretrainedonnx网络
`ExpientOnnXlayers`	Import layers fromonnx网络
`importONNXFunction`	导入预估计的ONNX网络作为功能

Parameters Imported by`importONNXFunction`

`onnxParameters`	导入的参数onnx网络for deep learning
`freezeParameters`	Convert learnable network parameters in`onnxParameters`to nonlearnable
`unfreezeParameters`	转换非可检测网络参数`onnxParameters`可以学习
`addParameter`	Add parameter to`onnxParameters`object
`删除参数`	Remove parameter from`onnxParameters`object

网络组件

`functionLayer`	功能层
`Findplaceholderlayers`	Find placeholder layers in network architecture imported from Keras oronnx
`replaceLayer`	Replace layer in layer graph
`汇编工作`	Assemble deep learning network from pretrained layers
`占位符`	Layer replacing an unsupported Keras oronnx层或不支持的功能金宝app`functionToLayerGraph`
`addLayers`	Add layers to layer graph
`removeLayers`	从图层中删除图层

Network Export

exportONNXNetwork Export network toonnx模型格式

Topics

Assemble Network from Pretrained Keras Layers
This example shows how to import the layers from a pretrained Keras network, replace the unsupported layers with custom layers, and assemble the layers into a network ready for prediction.
用功能层替换未支撑的金宝app凯拉斯层
This example shows how to import the layers from a pretrained Keras network, replace the unsupported layers with function layers, and assemble the layers into a network ready for prediction.
推理将n Between TensorFlow and Imported Networks for Image Classification
使用预验证的网络在TensorFlow中执行预测，使用网络将网络导入MATLABImportTensorFlownetwork，和then compare inference results between TensorFlow and MATLAB networks.
推理将n Between ONNX and Imported Networks for Image Classification
Perform prediction in ONNX with a pretrained network, import the network into MATLAB usingimportONNXNetwork，和then compare inference results between ONNX and MATLAB networks.
Classify Sequence of Images in Simulink with Imported TensorFlow Network
Import a pretrained TensorFlow network usingImportTensorFlownetwork，和then use the Predict block for image classification in Simulink^®.
Deploy Imported Network with MATLAB Compiler
Import Keras and ONNX pretrained networks and deploy the networks usingMATLAB Compiler™.
选择要导入ONNX预审慎网络的函数
Import an ONNX pretrained network usingimportONNXNetwork,ExpientOnnXlayers, orimportONNXFunction.