importCaffeNetwork
Import pretrained convolutional neural network models from Caffe
Description
imports a pretrained network from Caffe[1]. The function returns the pretrained network with the architecture specified by thenet
= importCaffeNetwork(protofile
,datafile
).prototxt
fileprotofile
and with network weights specified by the.caffemodel
filedatafile
.
This function requiresDeep Learning Toolbox™ Importer for Caffe Modelssupport package. If this support package is not installed, the function provides a download link.
You can download pretrained networks from Caffe Model Zoo[2].
returns a network with additional options specified by one or morenet
= importCaffeNetwork(___,Name,Value
)Name,Value
pair arguments using any of the previous syntaxes.
Examples
Input Arguments
Output Arguments
More About
Tips
importCaffeNetwork
can import networks with the following Caffe layer types, with some limitations:Caffe Layer Deep Learning Toolbox Layer BatchNormLayer
ConcatLayer
ConvolutionLayer
DeconvolutionLayer
DropoutLayer
EltwiseLayer
(only sum)EuclideanLossLayer
InnerProductLayer
InputLayer
LRNLayer
(Local Response Normalization)PoolingLayer
ReLULayer
ScaleLayer
SigmoidLayer
nnet.caffe.layer.SigmoidLayer
SoftmaxLayer
TanHLayer
If the network contains any other type of layer, then the software returns an error.
The function imports only the layers that
protofile
specifies with the include-phase TEST. The function ignores any layers thatprotofile
specifies with the include-phase TRAIN.MATLAB uses one-based indexing, whereas Python®uses zero-based indexing. In other words, the first element in an array has an index of 1 and 0 in MATLAB and Python, respectively. For more information on MATLAB indexing, seeArray Indexing. In MATLAB, to use an array of indices (
ind
) created in Python, convert the array toind+1
.
References
[1]Caffe.https://caffe.berkeleyvision.org/.
[2]Caffe Model Zoo.https://caffe.berkeleyvision.org/model_zoo.html.