sue602 / tiny-dnn Goto Github PK

tiny-dnn is a C++11 implementation of deep learning. It is suitable for deep learning on limited computational resource, embedded systems and IoT devices.

Linux/Mac OS	Windows

• Features
• Comparison with other libraries
• Supported networks
• Dependencies
• Build
• Examples
• Contributing
• References
• License
• Gitter rooms

Check out the documentation for more info.

• 2016/9/14 tiny-dnn v1.0.0alpha is released!
• 2016/8/7 tiny-dnn is now moved to organization account, and rename into tiny-dnn :)
• 2016/7/27 tiny-dnn v0.1.1 released!

• reasonably fast, without GPU
  • with TBB threading and SSE/AVX vectorization
  • 98.8% accuracy on MNIST in 13 minutes training (@Core i7-3520M)
• portable & header-only
  • Run anywhere as long as you have a compiler which supports C++11
  • Just include tiny_dnn.h and write your model in C++. There is nothing to install.
• easy to integrate with real applications
  • no output to stdout/stderr
  • a constant throughput (simple parallelization model, no garbage collection)
  • work without throwing an exception
  • can import caffe's model
• simply implemented
  • be a good library for learning neural networks

*unofficial version is available

• core
  • fully-connected
  • dropout
  • linear operation
  • power
• convolution
  • convolutional
  • average pooling
  • max pooling
  • deconvolutional
  • average unpooling
  • max unpooling
• normalization
  • contrast normalization (only forward pass)
  • batch normalization
• split/merge
  • concat
  • slice
  • elementwise-add

• tanh
• sigmoid
• softmax
• rectified linear(relu)
• leaky relu
• identity
• exponential linear units(elu)

• cross-entropy
• mean squared error
• mean absolute error
• mean absolute error with epsilon range

• stochastic gradient descent (with/without L2 normalization and momentum)
• adagrad
• rmsprop
• adam

Nothing. All you need is a C++11 compiler.

tiny-dnn is header-ony, so there's nothing to build. If you want to execute sample program or unit tests, you need to install cmake and type the following commands:

cmake .

Then open .sln file in visual studio and build(on windows/msvc), or type make command(on linux/mac/windows-mingw).

Some cmake options are available:

¹ tiny-dnn use c++11 standard library for parallelization by default

² If you don't use serialization, you can switch off to speedup compilation time.

³ tiny-dnn uses Google Test as default framework to run unit tests. No pre-installation required, it's automatically downloaded during CMake configuration.

For example, type the following commands if you want to use intel TBB and build tests:

cmake -DUSE_TBB=ON -DBUILD_TESTS=ON .

You can edit include/config.h to customize default behavior.

construct convolutional neural networks

#include "tiny_dnn/tiny_dnn.h"
using namespace tiny_dnn;
using namespace tiny_dnn::activation;
using namespace tiny_dnn::layers;

void construct_cnn() {
    using namespace tiny_dnn;

    network<sequential> net;

    // add layers
    net << conv<tan_h>(32, 32, 5, 1, 6)  // in:32x32x1, 5x5conv, 6fmaps
        << ave_pool<tan_h>(28, 28, 6, 2) // in:28x28x6, 2x2pooling
        << fc<tan_h>(14 * 14 * 6, 120)   // in:14x14x6, out:120
        << fc<identity>(120, 10);        // in:120,     out:10

    assert(net.in_data_size() == 32 * 32);
    assert(net.out_data_size() == 10);

    // load MNIST dataset
    std::vector<label_t> train_labels;
    std::vector<vec_t> train_images;

    parse_mnist_labels("train-labels.idx1-ubyte", &train_labels);
    parse_mnist_images("train-images.idx3-ubyte", &train_images, -1.0, 1.0, 2, 2);

    // declare optimization algorithm
    adagrad optimizer;

    // train (50-epoch, 30-minibatch)
    net.train<mse>(optimizer, train_images, train_labels, 30, 50);

    // save
    net.save("net");

    // load
    // network<sequential> net2;
    // net2.load("net");
}

construct multi-layer perceptron(mlp)

#include "tiny_dnn/tiny_dnn.h"
using namespace tiny_dnn;
using namespace tiny_dnn::activation;
using namespace tiny_dnn::layers;

void construct_mlp() {
    network<sequential> net;

    net << fc<sigmoid>(32 * 32, 300)
        << fc<identity>(300, 10);

    assert(net.in_data_size() == 32 * 32);
    assert(net.out_data_size() == 10);
}

another way to construct mlp

#include "tiny_dnn/tiny_dnn.h"
using namespace tiny_dnn;
using namespace tiny_dnn::activation;

void construct_mlp() {
    auto mynet = make_mlp<tan_h>({ 32 * 32, 300, 10 });

    assert(mynet.in_data_size() == 32 * 32);
    assert(mynet.out_data_size() == 10);
}

more sample, read examples/main.cpp or MNIST example page.

Since deep learning community is rapidly growing, we'd love to get contributions from you to accelerate tiny-dnn development! For a quick guide to contributing, take a look at the Contribution Documents.

[1] Y. Bengio, Practical Recommendations for Gradient-Based Training of Deep Architectures. arXiv:1206.5533v2, 2012

[2] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86, 2278-2324.

other useful reference lists:

• UFLDL Recommended Readings
• deeplearning.net reading list

The BSD 3-Clause License

We have a gitter rooms for discussing new features & QA. Feel free to join us!