# -*- coding: utf-8 -*-
#
# Copyright 2018-2020 Data61, CSIRO
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import tensorflow as tf
from tensorflow.keras.layers import Layer
from ..core.experimental import experimental
from ..core.validation import require_integer_in_range
[docs]class SortPooling(Layer):
"""
Sort Pooling Keras layer.
Note that sorting is performed using only the last column of the input tensor as stated in [1], "For convenience,
we set the last graph convolution to have one channel and only used this single channel for sorting."
[1] An End-to-End Deep Learning Atchitecture for Graph Classification, M. Zhang, Z. Cui, M. Neumann, and
Y. Chen, AAAI-18, https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/viewPaper/17146
Args:
k (int): The number of rows of output tensor.
flatten_output (bool): If True then the output tensor is reshaped to vector for each element in the batch.
"""
def __init__(self, k, flatten_output=False):
super().__init__()
require_integer_in_range(k, "k", min_val=1)
self.trainable = False
self.k = k
self.flatten_output = flatten_output
[docs] def get_config(self):
"""
Gets class configuration for Keras serialization. Used by keras model serialization.
Returns:
A dictionary that contains the config of the layer
"""
return {"k": self.k, "flatten_output": self.flatten_output}
[docs] def compute_output_shape(self, input_shapes):
"""
Computes the output shape of the layer.
Assumes the following inputs:
Args:
input_shapes (tuple of ints)
Shape tuples can include None for free dimensions, instead of an integer.
Returns:
An input shape tuple.
"""
if self.flatten_output:
return input_shapes[0], self.k * input_shapes[2], 1
else:
return input_shapes[0], self.k, input_shapes[2]
def _sort_tensor_with_mask(self, inputs):
embeddings, mask = inputs[0], inputs[1]
masked_sorted_embeddings = tf.gather(
embeddings,
tf.argsort(
tf.boolean_mask(embeddings, mask)[..., -1],
axis=0,
direction="DESCENDING",
),
)
embeddings = tf.pad(
masked_sorted_embeddings,
[
[0, (tf.shape(embeddings)[0] - tf.shape(masked_sorted_embeddings)[0])],
[0, 0],
],
)
return embeddings
[docs] def call(self, embeddings, mask):
"""
Applies the layer.
Args:
embeddings (tensor): the node features (size B x N x Sum F_i)
where B is the batch size, N is the number of nodes in the largest graph in the batch, and
F_i is the dimensionality of node features output from the i-th convolutional layer.
mask (tensor): a boolean mask (size B x N)
Returns:
Keras Tensor that represents the output of the layer.
"""
outputs = tf.map_fn(
self._sort_tensor_with_mask, (embeddings, mask), dtype=embeddings.dtype
)
# padding or truncation based on the value of self.k and the graph size (number of nodes)
outputs_shape = tf.shape(outputs)
outputs = tf.cond(
tf.math.less(outputs_shape, self.k)[1],
true_fn=lambda: tf.pad(
outputs, [[0, 0], [0, (self.k - outputs_shape)[1]], [0, 0]]
),
false_fn=lambda: outputs[:, : self.k, :],
)
if self.flatten_output:
outputs = tf.reshape(
outputs, [outputs_shape[0], embeddings.shape[-1] * self.k, 1]
)
return outputs