Legacy version of CNN neural net toolkit (now called dynet)
C++ Python Jupyter Notebook CMake Cuda Makefile
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.
pyexamples Changed attention function Jun 20, 2016



Legacy version of CNN neural net toolkit (now called dynet)


C++ neural network library

Important: Eigen version requirement

You need the development version of the Eigen library for this software to function. If you use any of the released versions, you may get assertion failures or compile errors.


First you need to fetch the dependent libraries

git submodule init
git submodule update

In src, you need to first use cmake to generate the makefiles

mkdir build
cd build
cmake .. -DEIGEN3_INCLUDE_DIR=/path/to/eigen

Then to compile, run

make -j 2

To see that things have built properly, you can run


which will train a multilayer perceptron to predict the xor function.

Building without Eigen installed

If you don't have Eigen installed, the instructions below will fetch and compile both Eigen and cnn. Eigen does not have to be compiled, so “installing” it is easy.

git clone https://github.com/clab/cnn.git
hg clone https://bitbucket.org/eigen/eigen/

cd cnn/
mkdir build
cd build
cmake .. -DEIGEN3_INCLUDE_DIR=../eigen
make -j 2

Debugging build problems

If you want to see the compile commands that are used, you can run

make VERBOSE=1

Training Models

An illustation of how models are trained (for a simple logistic regression model) is below:

// *** First, we set up the structure of the model
// Create a model, and an SGD trainer to update its parameters.
Model mod;
SimpleSGDTrainer sgd(&mod);
// Create a "computation graph," which will define the flow of information.
ComputationGraph cg;
// Initialize a 1x3 parameter vector, and add the parameters to be part of the
// computation graph.
Expression W = parameter(cg, mod.add_parameters({1, 3}));
// Create variables defining the input and output of the regression, and load them
// into the computation graph. Note that we don't need to set concrete values yet.
vector<cnn::real> x_values(3);
Expression x = input(cg, {3}, &x_values);
cnn::real y_value;
Expression y = input(cg, &y_value);
// Next, set up the structure to multiply the input by the weight vector,  then run
// the output of this through a logistic sigmoid function (logistic regression).
Expression y_pred = logistic(W*x);
// Finally, we create a function to calculate the loss. The model will be optimized
// to minimize the value of the final function in the computation graph.
Expression l = binary_log_loss(y_pred, y);
// We are now done setting up the graph, and we can print out its structure:

// *** Now, we perform a parameter update for a single example.
// Set the input/output to the values specified by the training data:
x_values = {0.5, 0.3, 0.7};
y_value = 1.0;
// "forward" propagates values forward through the computation graph, and returns
// the loss.
cnn::real loss = as_scalar(cg.forward());
// "backward" performs back-propagation, and accumulates the gradients of the
// parameters within the "Model" data structure.
// "sgd.update" updates parameters of the model that was passed to its constructor.
// Here 1.0 is the scaling factor that allows us to control the size of the update.

Note that this very simple example that doesn't cover things like memory initialization, reading/writing models, recurrent/LSTM networks, or adding biases to functions. The best way to get an idea of how to use cnn for real is to look in the example directory, particularly starting with the simplest xor example.