Simple feedforward neural network class written in pure python+numpy
Jupyter Notebook Python
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.


Simple feedforward neural network class "handmade" in pure python+numpy
Winner of Siraj Ravel's coding challange


  • Fully connected feed forward neural network with as many hidden layers as needed, of any size.
  • Implements backpropagation : link 1, link 2, link 3 (does not implement gradient checking)
  • Uses gradient descent for optimization
  • Uses Sum Squared Error cost function
  • Uses the sigmoid activation function

How to use

  1. Create an instance
    nn = NeuralNetwork()
  2. Add layers
    The first input layer is created automatically
    nn.add_layer(shape=(input_dim1, output_dim1))
    nn.add_layer(shape=(input_dim2, output_dim2))
  3. Train
    nn.train(features, targets, num_epochs, learning_rate)
  4. Predict

The nn.train() method has an optional parameter called stop_accuracy. At the end of each epoch the mean loss is calculated and if it is under the specified threshold than the training stops. This avoids training longer than necessary. By looking at the number of epochs needed to reach the threshold, it gives us a good metric as to the performance of our hyperparameters.

Example: XOR function

The Jupyter Notebook shows how the network can be used to approximate the XOR function using a 3-layer neural network. We attempt to find the optimal network dimensions and the optimal learning rate.

Error Optimization


  • Numpy