EfficientNet Trained on ImageNet with NoisyStudent

Identify the main object in an image

Released in 2019, this model utilizes the techniques of NoisyStudent data augmentation on the EfficientNet architectures to effectively perform image classification.

Number of models: 8

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

The models achieve the following accuracies on the original ImageNet validation set.

Examples

Download Example Notebook

Open in Wolfram Cloud

Resource retrieval

Get the pre-trained net:

In[1]:=

$NetModel["EfficientNet Trained on ImageNet with NoisyStudent", \ "ParametersInformation"]$

Out[2]=

Out[4]=

Entity object, which can be queried:

In[6]:=

Out[7]=

(* Evaluate this cell to get the example input *) CloudGet["https://www.wolframcloud.com/obj/001f5dae-ed4e-4140-989a-c85cc38f5411"]

Out[9]=

EntityValue[
NetExtract[
NetModel["EfficientNet Trained on ImageNet with NoisyStudent"], "Output"][["Labels"]], "Name"]

Out[10]=

Out[11]=

Visualize convolutional weights

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

In[14]:=

$weights = NetExtract[ NetModel[ "EfficientNet Trained on ImageNet with NoisyStudent"], \ {"stem_conv", "Weights"}];$

Show the dimensions of the weights:

In[16]:=

Transfer learning

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

In[18]:=

Out[20]=

TargetDevice -> "GPU" for training on a GPU):

In[22]:=

Net information

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

In[24]:=

$Information[ NetModel["EfficientNet Trained on ImageNet with NoisyStudent"], \ "ArraysElementCounts"]$

Out[24]=

$Information[ NetModel["EfficientNet Trained on ImageNet with NoisyStudent"], \ "ArraysTotalElementCount"]$

Out[25]=

$Information[ NetModel["EfficientNet Trained on ImageNet with NoisyStudent"], \ "LayerTypeCounts"]$

Out[26]=

Out[29]=

Construction Notebook

Download Construction Notebook

Open in Wolfram Cloud

Requirements

Wolfram Language 12.1 (March 2020) or above

Resource History

Date Created: 5 November 2020

Reference

M. Tan, Q. Le, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks," arXiv:1905.11946 (2019)
Q. Xie, M.-T. Luong, E. Hovy, Q. Le, "Self-Training with Noisy Student Improves ImageNet Classification," arXiv:1911.04252 (2019)
Available from: https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet