1. Advisor
  2. anna
  3. functions
  4. anna.layers.layers.DenseLayer
View all anna analysis

How to use the anna.layers.layers.DenseLayer function in anna

To help you get started, we’ve selected a few anna examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

github ifp-uiuc / an-analysis-of-unsupervised-pre-training-iclr-2015 / stl10 / models.py View on Github external
nonlinearity=nonlinearity,
        pad=2)
    pool3 = cc_layers.Pooling2DLayer(conv3, 12, stride=12)

    winitD1 = k/numpy.sqrt(numpy.prod(pool3.get_output_shape()))
    winitD2 = k/numpy.sqrt(512)

    pool3_shuffle = cc_layers.ShuffleC01BToBC01Layer(pool3)
    fc4 = layers.DenseLayer(
        pool3_shuffle,
        n_outputs=512,
        weights_std=winitD1,
        init_bias_value=1.0,
        nonlinearity=layers.rectify,
        dropout=0.5)
    output = layers.DenseLayer(
        fc4,
        n_outputs=10,
        weights_std=winitD2,
        init_bias_value=0.0,
        nonlinearity=layers.softmax)
github ifp-uiuc / an-analysis-of-unsupervised-pre-training-iclr-2015 / stl10 / cnn_adcu / model.py View on Github external
pad=2)
    pool2 = cc_layers.Pooling2DLayer(conv2, 2, stride=2)
    conv3 = cc_layers.Conv2DNoBiasLayer(
        pool2,
        n_filters=256,
        filter_size=5,
        weights_std=winit3,
        nonlinearity=nonlinearity,
        pad=2)
    pool3 = cc_layers.Pooling2DLayer(conv3, 12, stride=12)

    winitD1 = k/numpy.sqrt(numpy.prod(pool3.get_output_shape()))
    winitD2 = k/numpy.sqrt(512)

    pool3_shuffle = cc_layers.ShuffleC01BToBC01Layer(pool3)
    fc4 = layers.DenseLayer(
        pool3_shuffle,
        n_outputs=512,
        weights_std=winitD1,
        init_bias_value=1.0,
        nonlinearity=layers.rectify,
        dropout=0.5)
    output = layers.DenseLayer(
        fc4,
        n_outputs=10,
        weights_std=winitD2,
        init_bias_value=0.0,
        nonlinearity=layers.softmax)
github ifp-uiuc / an-analysis-of-unsupervised-pre-training-iclr-2015 / stl10 / models.py View on Github external
pad=2)
    pool2 = cc_layers.Pooling2DLayer(conv2, 2, stride=2)
    conv3 = cc_layers.Conv2DNoBiasLayer(
        pool2,
        n_filters=256,
        filter_size=5,
        weights_std=winit3,
        nonlinearity=nonlinearity,
        pad=2)
    pool3 = cc_layers.Pooling2DLayer(conv3, 12, stride=12)

    winitD1 = k/numpy.sqrt(numpy.prod(pool3.get_output_shape()))
    winitD2 = k/numpy.sqrt(512)

    pool3_shuffle = cc_layers.ShuffleC01BToBC01Layer(pool3)
    fc4 = layers.DenseLayer(
        pool3_shuffle,
        n_outputs=512,
        weights_std=winitD1,
        init_bias_value=1.0,
        nonlinearity=layers.rectify,
        dropout=0.5)
    output = layers.DenseLayer(
        fc4,
        n_outputs=10,
        weights_std=winitD2,
        init_bias_value=0.0,
        nonlinearity=layers.softmax)
github ifp-uiuc / an-analysis-of-unsupervised-pre-training-iclr-2015 / stl10 / adu_model.py View on Github external
pad=2)
    pool2 = cc_layers.Pooling2DLayer(conv2, 2, stride=2)
    conv3 = cc_layers.Conv2DNoBiasLayer(
        pool2,
        n_filters=256,
        filter_size=5,
        weights_std=winit3,
        nonlinearity=nonlinearity,
        pad=2)
    pool3 = cc_layers.Pooling2DLayer(conv3, 12, stride=12)

    winitD1 = k/numpy.sqrt(numpy.prod(pool3.get_output_shape()))
    winitD2 = k/numpy.sqrt(512)

    pool3_shuffle = cc_layers.ShuffleC01BToBC01Layer(pool3)
    fc4 = layers.DenseLayer(
        pool3_shuffle,
        n_outputs=512,
        weights_std=winitD1,
        init_bias_value=1.0,
        nonlinearity=layers.rectify,
        dropout=0.5)
    output = layers.DenseLayer(
        fc4,
        n_outputs=10,
        weights_std=winitD2,
        init_bias_value=0.0,
        nonlinearity=layers.softmax)
github ifp-uiuc / an-analysis-of-unsupervised-pre-training-iclr-2015 / cifar10 / models.py View on Github external
nonlinearity=nonlinearity,
        pad=2)
    pool2 = cc_layers.Pooling2DLayer(conv2, 2, stride=2)
    conv3 = cc_layers.Conv2DNoBiasLayer(
        pool2,
        n_filters=192,
        filter_size=3,
        weights_std=winit3,
        nonlinearity=nonlinearity,
        pad=1)

    winitD1 = k/numpy.sqrt(numpy.prod(conv3.get_output_shape()))
    winitD2 = k/numpy.sqrt(300)

    conv3_shuffle = cc_layers.ShuffleC01BToBC01Layer(conv3)
    fc4 = layers.DenseLayer(
        conv3_shuffle,
        n_outputs=300,
        weights_std=winitD1,
        init_bias_value=1.0,
        nonlinearity=layers.rectify,
        dropout=0.0)
    output = layers.DenseLayer(
        fc4,
        n_outputs=10,
        weights_std=winitD2,
        init_bias_value=0.0,
        nonlinearity=layers.softmax)
github ifp-uiuc / an-analysis-of-unsupervised-pre-training-iclr-2015 / stl10 / cnn_adcu / model.py View on Github external
nonlinearity=nonlinearity,
        pad=2)
    pool3 = cc_layers.Pooling2DLayer(conv3, 12, stride=12)

    winitD1 = k/numpy.sqrt(numpy.prod(pool3.get_output_shape()))
    winitD2 = k/numpy.sqrt(512)

    pool3_shuffle = cc_layers.ShuffleC01BToBC01Layer(pool3)
    fc4 = layers.DenseLayer(
        pool3_shuffle,
        n_outputs=512,
        weights_std=winitD1,
        init_bias_value=1.0,
        nonlinearity=layers.rectify,
        dropout=0.5)
    output = layers.DenseLayer(
        fc4,
        n_outputs=10,
        weights_std=winitD2,
        init_bias_value=0.0,
        nonlinearity=layers.softmax)
github ifp-uiuc / an-analysis-of-unsupervised-pre-training-iclr-2015 / stl10 / cnn_adu / model.py View on Github external
pad=2)
    pool2 = cc_layers.CudaConvnetPooling2DLayer(conv2, 2, stride=2)
    conv3 = cc_layers.CudaConvnetConv2DNoBiasLayer(
        pool2, 
        n_filters=256,
        filter_size=5,
        weights_std=winit3,
        nonlinearity=nonlinearity,
        pad=2)   
    pool3 = cc_layers.CudaConvnetPooling2DLayer(conv3, 12, stride=12)

    winitD1 = k/numpy.sqrt(numpy.prod(pool3.get_output_shape()))
    winitD2 = k/numpy.sqrt(512)

    pool3_shuffle = cc_layers.ShuffleC01BToBC01Layer(pool3)    
    fc4 = layers.DenseLayer(
        pool3_shuffle,
        n_outputs = 512,
        weights_std=winitD1,
        init_bias_value=1.0,
        nonlinearity=layers.rectify,
        dropout=0.5)
    output = layers.DenseLayer(
        fc4,
        n_outputs=10,
        weights_std=winitD2,
        init_bias_value=0.0,
        nonlinearity=layers.softmax)

anna

A Neat configuratioN Auxiliary

BSD-3-Clause
Latest version published 4 months ago

Package Health Score

48 / 100
Full package analysis

Popular anna functions

Similar packages