ResNet-152 Trained on ImageNet Competition Data

Identify the main object in an image

Number of layers: 517 | Parameter count: 60,344,232 | Trained size: 244 MB |

Training Set Information

ImageNet Large Scale Visual Recognition Challenge 2012 classification dataset, consisting of 1.2 million training images, with 1,000 classes of objects.

Performance

This model achieves 77% top-1 and 93.3% top-5 accuracy in 1-crop validation, and 78.6% top-1 and 94.3% top-5 accuracy in 10-crop validation on the ImageNet Large Scale Visual Recognition Challenge 2012 dataset.

Examples

Download Example Notebook

Open in Wolfram Cloud

Resource retrieval

Get the pre-trained net:

In[1]:=

Out[1]=

Basic usage

Classify an image:

In[2]:=

(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/eec4deed-6fad-47a8-a8ee-73ed7c43d425"]

Out[2]=

The prediction is an Entity object, which can be queried:

In[3]:=

Out[3]=

Get a list of available properties of the predicted Entity:

In[4]:=

Out[4]=

Obtain the probabilities of the ten most likely entities predicted by the net:

In[5]:=

(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/f73e1d3b-580a-4ad1-b233-ea061c36e244"]

Out[5]=

An object outside the list of the ImageNet classes will be misidentified:

In[6]:=

(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/816302fc-46c6-44b5-9891-a43abe3e04b2"]

Out[6]=

Obtain the list of names of all available classes:

In[7]:=

EntityValue[
NetExtract[
NetModel["ResNet-152 Trained on ImageNet Competition Data"], "Output"][["Labels"]], "Name"]

Out[7]=

Feature extraction

Remove the last three layers of the trained net so that the net produces a vector representation of an image:

In[8]:=

Out[8]=

Get a set of images:

In[9]:=

(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/35170516-a42e-4df3-8db2-0292c69b6cff"]

Visualize the features of a set of images:

In[10]:=

Out[10]=

Visualize convolutional weights

Extract the weights of the first convolutional layer in the trained net:

In[11]:=

Visualize the weights as a list of 64 images of size 7x7:

In[12]:=

Out[12]=

Transfer learning

Use the pre-trained model to build a classifier for telling apart images of dogs and cats. Create a test set and a training set:

In[13]:=

(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/be97c07f-e5e1-44a6-8099-c5017d2d7e78"]

In[14]:=

(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/acdb90a5-4bd9-4018-b598-e3d1699c980f"]

Remove the linear layer from the pre-trained net:

In[15]:=

Out[15]=

Create a new net composed of the pre-trained net followed by a linear layer and a softmax layer:

In[16]:=

newNet = NetChain[<|"pretrainedNet" -> tempNet, "linearNew" -> LinearLayer[], "softmax" -> SoftmaxLayer[]|>, "Output" -> NetDecoder[{"Class", {"cat", "dog"}}]]

Out[16]=

Train on the dataset, freezing all the weights except for those in the "linearNew" layer (use TargetDevice -> "GPU" for training on a GPU):

In[17]:=

$trainedNet = NetTrain[newNet, trainSet, LearningRateMultipliers -> {"linearNew" -> 1, _ -> 0}]$

Out[17]=

Perfect accuracy is obtained on the test set:

In[18]:=

Out[18]=

Net information

Inspect the number of parameters of all arrays in the net:

In[19]:=

$NetInformation[ NetModel["ResNet-152 Trained on ImageNet Competition Data"], \ "ArraysElementCounts"]$

Out[19]=

Obtain the total number of parameters:

In[20]:=

$NetInformation[ NetModel["ResNet-152 Trained on ImageNet Competition Data"], \ "ArraysTotalElementCount"]$

Out[20]=

Obtain the layer type counts:

In[21]:=

$NetInformation[ NetModel["ResNet-152 Trained on ImageNet Competition Data"], \ "LayerTypeCounts"]$

Out[21]=

Display the summary graphic:

In[22]:=

$NetInformation[ NetModel["ResNet-152 Trained on ImageNet Competition Data"], \ "SummaryGraphic"]$

Out[22]=

Export to MXNet

Export the net into a format that can be opened in MXNet:

In[23]:=

Out[23]=

Export also creates a net.params file containing parameters:

In[24]:=

Out[24]=

Get the size of the parameter file:

In[25]:=

Out[25]=

The size is similar to the byte count of the resource object:

In[26]:=

Out[26]=

Construction Notebook

Download Construction Notebook

Open in Wolfram Cloud

Requirements

Wolfram Language 11.2 (September 2017) or above

Resource History

Date Created: 17 July 2017
Latest Update: 21 June 2018

Reference

K. He, X. Zhang, S. Ren, J. Sun, "Deep Residual Learning for Image Recognition," arXiv:1512.03385 (2015)
Available from: https://github.com/KaimingHe/deep-residual-networks
Rights: MIT License