megnet.layers.readout.set2set module

class Set2Set(T=3, n_hidden=512, activation=None, activation_lstm='tanh', recurrent_activation='hard_sigmoid', kernel_initializer='glorot_uniform', recurrent_initializer='orthogonal', bias_initializer='zeros', use_bias=True, unit_forget_bias=True, kernel_regularizer=None, recurrent_regularizer=None, bias_regularizer=None, kernel_constraint=None, recurrent_constraint=None, bias_constraint=None, **kwargs)[source]

Bases: tensorflow.python.keras.engine.base_layer.Layer

For a set of vectors, the set2set neural network maps it to a single vector. The order invariance is acheived by a attention mechanism. See Vinyals, Oriol, Samy Bengio, and Manjunath Kudlur. “Order matters: Sequence to sequence for sets.” arXiv preprint arXiv:1511.06391 (2015).

Parameters
  • T – (int) recurrent step

  • n_hidden – (int) number of hidden units

  • activation – (str or object) activation function

  • activation_lstm – (str or object) activation function for lstm

  • recurrent_activation – (str or object) activation function for recurrent step

  • kernel_initializer – (str or object) initializer for kernel weights

  • recurrent_initializer – (str or object) initializer for recurrent weights

  • bias_initializer – (str or object) initializer for biases

  • use_bias – (bool) whether to use biases

  • unit_forget_bias – (bool) whether to use basis in forget gate

  • kernel_regularizer – (str or object) regularizer for kernel weights

  • recurrent_regularizer – (str or object) regularizer for recurrent weights

  • bias_regularizer – (str or object) regularizer for biases

  • kernel_constraint – (str or object) constraint for kernel weights

  • recurrent_constraint – (str or object) constraint for recurrent weights

  • bias_constraint – (str or object) constraint for biases

  • kwargs – other inputs for keras Layer class

build(input_shape)[source]

Creates the variables of the layer (optional, for subclass implementers).

This is a method that implementers of subclasses of Layer or Model can override if they need a state-creation step in-between layer instantiation and layer call.

This is typically used to create the weights of Layer subclasses.

Parameters

input_shape – Instance of TensorShape, or list of instances of TensorShape if the layer expects a list of inputs (one instance per input).

call(inputs, mask=None)[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)[source]

Computes the output shape of the layer.

If the layer has not been built, this method will call build on the layer. This assumes that the layer will later be used with inputs that match the input shape provided here.

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.

get_config()[source]

Returns the config of the layer.

A layer config is a Python dictionary (serializable) containing the configuration of a layer. The same layer can be reinstantiated later (without its trained weights) from this configuration.

The config of a layer does not include connectivity information, nor the layer class name. These are handled by Network (one layer of abstraction above).

Returns

Python dictionary.