Skip to content

Instantly share code, notes, and snippets.

@Tathagatd96

Tathagatd96/gist:424ba17fcb6321c453fb3059ff9d0f86

Created December 6, 2017 13:33
Show Gist options
  • Save Tathagatd96/424ba17fcb6321c453fb3059ff9d0f86 to your computer and use it in GitHub Desktop.
Save Tathagatd96/424ba17fcb6321c453fb3059ff9d0f86 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
gistfile1.txt
#Input layer:
net['data'] = lasagne.layers.InputLayer(data_size, input_var=input_var)
#Convolution + Pooling + Normalization
net['conv1'] = lasagne.layers.Conv2DLayer(net['data'], num_filters=6, filter_size=3)
net['pool1'] = lasagne.layers.Pool2DLayer(net['conv1'], pool_size=2)
net['conv2'] = lasagne.layers.Conv2DLayer(net['pool1'], num_filters=10, filter_size=4)
net['pool2'] = lasagne.layers.Pool2DLayer(net['conv2'], pool_size=2)
net['conv3'] = lasagne.layers.Conv2DLayer(net['pool2'], num_filters=20, filter_size=2)
net['conv4'] = lasagne.layers.Conv2DLayer(net['conv3'], num_filters=20, filter_size=2)
net['conv5'] = lasagne.layers.Conv2DLayer(net['conv4'], num_filters=20, filter_size=2)
net['pool3'] = lasagne.layers.Pool2DLayer(net['conv5'], pool_size=2)
#Fully-connected
net['fc1'] = lasagne.layers.DenseLayer(net['pool3'], num_units=100)
net['fc2'] = lasagne.layers.DenseLayer(net['fc1'], num_units=100)
#Output layer:
net['out'] = lasagne.layers.DenseLayer(net['fc2'], num_units=output_size,
nonlinearity=lasagne.nonlinearities.softmax)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment