Back to Cntk

CNTK Examples: Image/Classification/ConvNet

Examples/Image/Classification/ConvNet/README.md

2015-12-081.9 KB
Original Source

CNTK Examples: Image/Classification/ConvNet

Overview

Data:The MNIST dataset (http://yann.lecun.com/exdb/mnist/) of handwritten digits and the CIFAR-10 dataset (http://www.cs.toronto.edu/~kriz/cifar.html) for image classification.
PurposeThis folder contains a number of examples that demonstrate how to use CNTK to define convolutional neural networks for image classification.
NetworkConvolutional neural networks.
TrainingStochastic gradient descent with momentum.
CommentsSee below.

Running the example

Getting the data

We use the MNIST and CIFAR-10 datasets to demonstrate how to train a convolutional neural network (CNN). CNN has been one of the most popular neural networks for image-related tasks. A very well-known early work on CNN is the LeNet. In 2012 Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton won the ILSVRC-2012 competition using a CNN architecture, AlexNet. And most state-of-the-art neural networks on image classification tasks today adopt a modified CNN architecture, such as VGG, GoogLeNet, ResNet, etc.

MNIST and CIFAR-10 datasets are not included in the CNTK distribution but can be easily downloaded and converted by following the instructions in DataSets/MNIST and DataSets/CIFAR-10. We recommend you to keep the downloaded data in the respective folder while downloading, as the configuration files in this folder assumes that by default.

Details

We offer multiple CNN examples, including one for the MNIST dataset, and two for the CIFAR-10 dataset (one with and one without data augmentation). For details, please click the respective links below.

Python

BrainScript