twodlearn.feedforward module¶

class twodlearn.feedforward.AffineLayer(units, *args, **kargs)[source]¶

Bases: twodlearn.feedforward.LinearLayer

Standard affine (W*X+b) fully connected layer

Tdl autoinitialization with arguments:

kernel[source]¶: (ParameterInit) Autoinit with arguments [‘initializer’, ‘trainable’]

input_shape[source]¶: (InputArgument)

regularizer[source]¶: (Regularizer) Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

units[source]¶: (InputArgument) Number of output units (int).

bias[source]¶: (ParameterInit) Autoinit with arguments [‘initializer’, ‘trainable’]

class Output(model, inputs, options=None, name=None)[source]¶

Bases: twodlearn.feedforward.Output

property bias[source]¶

value[source]¶

bias[source]: Autoinit with arguments [‘initializer’, ‘trainable’]

class twodlearn.feedforward.AlexNet(input_shape, n_outputs, n_filters, filter_sizes, pool_sizes, n_hidden, output_function=None, name='AlexNet')[source]¶

Bases: twodlearn.core.common.TdlModel

class AlexNetSetup(model, inputs=None, batch_size=None, options=None, name='AlexNet')[source]¶

Bases: twodlearn.core.common.TdlModel

conv[source]¶

property input_shape[source]¶

inputs[source]¶

loss[source]¶: Decorator used to specify an optional property inside a model. The decorator works similar to @property, but the specified method correponds to the initialization of the property

mlp[source]¶

output[source]¶

property value[source]¶

property weights[source]¶

conv[source]¶

evaluate(inputs=None, options=None, name=None)[source]¶

property input_shape[source]¶

mlp[source]¶

property n_outputs[source]¶

regularizer[source]¶: Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

property weights[source]¶

class twodlearn.feedforward.AlexNetClassifier(input_shape, n_classes, n_filters, filter_sizes, pool_sizes, n_hidden, name='AlexNetClassifier')[source]¶

Bases: twodlearn.feedforward.AlexNet

class AlexNetOutput(model, inputs=None, batch_size=None, options=None, name='AlexNet')[source]¶

Bases: twodlearn.feedforward.AlexNetSetup

property logits[source]¶

loss[source]¶: Decorator used to specify an optional property inside a model. The decorator works similar to @property, but the specified method correponds to the initialization of the property

evaluate(inputs=None, options=None, name=None)[source]¶

mlp[source]¶

property n_classes[source]¶

class twodlearn.feedforward.AlexnetLayer(filter_size, n_maps, pool_size, name=None)[source]¶

Bases: twodlearn.core.common.TdlModel

Creates a layer like the one used in (ImageNet Classification with Deep Convolutional Neural Networks).

The format for filter_size is:: [filter_size_dim0 , filter_size_dim1], it performs 2D convolution
The format for n_maps is:: [num_input_maps, num_output_maps]
The format for pool_size is:: [pool_size_dim0, pool_size_dim1]

class AlexnetLayerSetup(**kargs)[source]¶

Bases: twodlearn.core.common.TdlModel

property bias[source]¶

eval_layer(inputs)[source]¶

inputs[source]¶

model[source]¶

property pool_size[source]¶

value[source]¶

property weights[source]¶

property y[source]¶

bias[source]¶

evaluate(inputs, name=None)[source]¶

property filter_size[source]¶

property n_maps[source]¶

property pool_size[source]¶

regularizer[source]¶: Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

weights[source]¶

class twodlearn.feedforward.BoundedOutput(lower=1e-07, upper=None, name='BoundedOutput')[source]¶

Bases: twodlearn.core.common.TdlModel

class Output(**kargs)[source]¶

Bases: twodlearn.core.common.OutputModel

property lower[source]¶

property upper[source]¶

evaluate(*args, **kargs)[source]¶

lower[source]¶

upper[source]¶

class twodlearn.feedforward.Concat(axis, name='Concat')[source]¶

Bases: twodlearn.core.common.TdlModel

property axis[source]¶

evaluate(*args, **kargs)[source]¶

class twodlearn.feedforward.DenseLayer(activation=<function relu>, name=None, **kargs)[source]¶

Bases: twodlearn.feedforward.AffineLayer

Standard fully connected layer

Tdl autoinitialization with arguments:

kernel[source]¶: (ParameterInit) Autoinit with arguments [‘initializer’, ‘trainable’]

input_shape[source]¶: (InputArgument)

regularizer[source]¶: (Regularizer) Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

activation[source]¶: (InputArgument)

units[source]¶: (InputArgument) Number of output units (int).

bias[source]¶: (ParameterInit) Autoinit with arguments [‘initializer’, ‘trainable’]

class Output(model, inputs, options=None, name=None)[source]¶

Bases: twodlearn.feedforward.Output

property activation[source]¶

property affine[source]¶: activation before non-linearity

value[source]¶

activation[source]

class twodlearn.feedforward.LinearClassifier(n_inputs, n_classes, name='linear_classifier', **kargs)[source]¶

Bases: twodlearn.core.common.TdlModel

class LinearClassifierSetup(**kargs)[source]¶

Bases: twodlearn.core.common.OutputModel

property labels[source]¶

loss[source]¶: Decorator used to specify an optional property inside a model. The decorator works similar to @property, but the specified method correponds to the initialization of the property

property n_inputs[source]¶

property n_outputs[source]¶

property weights[source]¶

evaluate(*args, **kargs)[source]¶

linear_layer[source]¶

property n_classes[source]¶

property n_inputs[source]¶

property n_outputs[source]¶

regularizer[source]¶: Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

property weights[source]¶

class twodlearn.feedforward.LinearLayer(units, *args, **kargs)[source]¶

Bases: twodlearn.core.layers.Layer

Standard linear (W*X) fully connected layer

Tdl autoinitialization with arguments:

kernel[source]¶: (ParameterInit) Autoinit with arguments [‘initializer’, ‘trainable’]

input_shape[source]¶: (InputArgument)

units[source]¶: (InputArgument) Number of output units (int).

regularizer[source]¶: (Regularizer) Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

class Output(model, inputs, options=None, name=None)[source]¶

Bases: twodlearn.core.common.TdlModel

inputs[source]¶

property kernel[source]¶

property shape[source]¶

value[source]¶

call(inputs, *args, **kargs)[source]¶

This is where the layer’s logic lives.

Parameters

inputs – Input tensor, or list/tuple of input tensors.
**kwargs – Additional keyword arguments.

Returns

A tensor or list/tuple of tensors.

compute_output_shape(input_shape=None)[source]¶

Computes the output shape of the layer.

Assumes that the layer will be built to match that input shape provided.

Parameters: input_shape – Shape tuple (tuple of integers) or list of shape tuples (one per output tensor of the layer). Shape tuples can include None for free dimensions, instead of an integer.
Returns: An input shape tuple.

input_shape[source]

kernel[source]: Autoinit with arguments [‘initializer’, ‘trainable’]

regularizer[source]: Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

units[source]: Number of output units (int).

class twodlearn.feedforward.MlpClassifier(n_inputs, n_classes, n_hidden, afunction=<function relu>, name=None)[source]¶

Bases: twodlearn.feedforward.MlpNet

class Output(model, inputs=None, keep_prob=None, name=None)[source]¶

Bases: twodlearn.feedforward.Output

property logits[source]¶

loss[source]¶: Decorator used to specify an optional property inside a model. The decorator works similar to @property, but the specified method correponds to the initialization of the property

evaluate(inputs=None, keep_prob=None, name=None)[source]¶

property n_classes[source]¶

class twodlearn.feedforward.MlpNet(n_inputs, n_outputs, n_hidden, afunction=<function relu>, output_function=None, name='MlpNet')[source]¶

Bases: twodlearn.feedforward.StackedModel

full_layers: list of fully connected layers out_layer: output layer, for the moment, linear layer

class Output(model, inputs=None, keep_prob=None, name=None)[source]¶

Bases: twodlearn.core.common.TdlModel

hidden[source]¶

inputs[source]¶

keep_prob[source]¶: probability of not droping an activation during dropout

loss[source]¶: Decorator used to specify an optional property inside a model. The decorator works similar to @property, but the specified method correponds to the initialization of the property

property n_inputs[source]¶

property n_outputs[source]¶

output[source]¶: output from the network, after output_function

property shape[source]¶

value[source]¶

property weights[source]¶

add(layer)[source]¶

evaluate(inputs=None, keep_prob=None, name=None)[source]¶

layers[source]¶

property n_hidden[source]¶

property n_inputs[source]¶

property n_outputs[source]¶

property weights[source]¶

class twodlearn.feedforward.MultiLayer2DConvolution(input_shape, n_filters, filter_sizes, pool_sizes, name='MultiConv2D')[source]¶

Bases: twodlearn.core.common.TdlModel

Creates a Convolutional neural network

It performs a series of 2d Convolutions and pooling operations

input_size: size of the input maps, [size_dim0, size_dim1] n_outputs: number of outputs n_input_maps: number of input maps n_filters: list with the number of filters for layer filter_size: list with the size of the kernel for each layer,

the format for the size of each layer is: [filter_size_dim0 , filter_size_dim1]

pool_size: list with the size of the pooling kernel foreach layer,: the format for each layer is: [pool_size_dim0, pool_size_dim1]

class Output(model, inputs=None, batch_size=None, options=None, name='MultiConv2D')[source]¶

Bases: twodlearn.core.common.TdlModel

property hidden[source]¶

property input_shape[source]¶

inputs[source]¶

model[source]¶

setup_conv_layers(inputs, layers)[source]¶

setup_inputs(batch_size, input_shape)[source]¶

value[source]¶

property weights[source]¶

property y[source]¶

property filter_sizes[source]¶

property input_shape[source]¶

layers[source]¶

property n_filters[source]¶

property output_shape[source]¶

property pool_sizes[source]¶

regularizer[source]¶: Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

setup(inputs=None, batch_size=None, options=None, name=None)[source]¶

property weights[source]¶

class twodlearn.feedforward.NetConf(inputs, labels, y, loss)[source]¶

Bases: object

This is a wrapper to any network configuration, it contains the references to the placeholders for inputs and labels, and the reference of the computation graph for the network

inputs: placeholder for the inputs labels: placeholder for the labels y: output of the comptuation graph, ussually a linear map

from the last layer (logits)

loss: loss for the network

class twodlearn.feedforward.Options(weight_initialization, weight_initialization_alpha)[source]¶: Bases: object

class twodlearn.feedforward.ParallelModel(models, name='Parallel')[source]¶

Bases: twodlearn.core.common.TdlModel

evaluate(*args, **kargs)[source]¶

property models[source]¶

class twodlearn.feedforward.StackedModel(layers=None, return_layers=None, options=None, name='Stacked')[source]¶

Bases: twodlearn.core.common.TdlModel

class StackedOutput(**kargs)[source]¶: Bases: twodlearn.core.common.OutputModel

add(layer, name=None)[source]¶

evaluate(*args, **kargs)[source]¶

get_save_data()[source]¶

layers[source]¶

regularizer[source]¶: Decorator used to specify a regularizer for a model. The decorator works similar to @property, but the specified method correponds to the initialization of the regularizer.

return_layers[source]¶: True if the return value of the stacked model is the layers

class twodlearn.feedforward.StridedDeconvNet(n_inputs, input_size, n_input_maps, n_filters, filter_size, upsampling, name='')[source]¶

Bases: object

Creates a Deconvolutional neural network using upsampling TODO: implement this using new format It performs a ‘deconvolutional’ neural network similar to the one used in “UNSUPERVISED REPRESENTATION LEARNING WITH DEEP CONVOLUTIONAL GENERATIVE ADVERSARIAL NETWORKS” (http://arxiv.org/pdf/1511.06434v2.pdf)

The network maps a vector of size n_inputs to a 2d map with several chanels

First a linear mapping is performed, then a reshape to form an initial tensor of 2d maps with chanels, then a series of upscaling and convolutions are performed

n_inputs: size of the input vectors input_size: size of the maps after linear stage: [size_dim0, size_dim1] n_input_maps: number of maps after linear stage n_filters: list with the number of filters for each layer filter_size: list with the size of the kernel for each layer,

the format for the size of each layer is: [filter_size_dim0 , filter_size_dim1]

upsampling: list with the size for the upsampling in each deconv layer:: [upsampling_dim0, upsampling_dim1]
in_layer: input layer, a linear layer for mapping the inputs to the desired: output

setup(batch_size, drop_prob=None)[source]¶

Defines the computation graph of the neural network for a specific: batch size
drop_prob: placeholder used for specify the probability for dropout. If: this coefficient is set, then dropout regularization is added between all fully connected layers(TODO: allow to choose which layers)

class twodlearn.feedforward.Transpose(**kargs)[source]¶

Bases: twodlearn.core.common.TdlModel

inputs[source]¶

perm[source]¶

rightmost[source]¶

shape[source]¶

value[source]¶

class twodlearn.feedforward.TransposeLayer(**kargs)[source]¶

Bases: twodlearn.core.common.TdlModel

perm[source]¶: Autoinit with arguments [‘inputs’]

rightmost[source]¶

twodlearn.feedforward.leaky_relu(x, leaky_slope=0.01)[source]¶: leaky relu, with 0.01 slope for negative values

twodlearn.feedforward.options = <twodlearn.feedforward.Options object>[source]¶: ————————- Activation functions ————————

twodlearn.feedforward.selu01(x)[source]¶: Self normalizing activation function Activation function proposed by Gunter Klambauer et. al. “Self-Normalizing Neural Networks”, https://arxiv.org/abs/1706.02515

twodlearn.feedforward.selu01_disc(x)[source]¶: discontinuous selu

twodlearn.feedforward.selu01_disc2(x)[source]¶: another version of discontinuous selu