Deep Semantic Structured Model (DSSM)

The module dssm provides functions to build a DSSM.

dssm.build_multi_dssm(input_var=None, num_samples=None, num_entries=6, num_ngrams=74088, num_hid1=300, num_hid2=300, num_out=128)[source]

Builds a DSSM structure in a Lasagne/Theano way.

The built DSSM is the neural network that computes the projection of only one paper. The input input_var should have two dimensions: (num_samples * num_entries, num_ngrams). The output is then computed in a batch way: one paper at a time, but all papers from the same sample in the dataset are grouped (cited papers, citing papers and num_entries - 2 irrelevant papers).

Parameters:
  • input_var (theano.tensor.TensorType or None) – symbolic input variable of the DSSM
  • num_samples (int) – the number of samples in the batch input dataset (number of rows)
  • num_entries (int) – the number of compared papers in the DSSM structure
  • num_ngrams (int) – the size of the vocabulary
  • num_hid1 (int) – the number of units in the first hidden layer
  • num_hid2 (int) – the number of units in the second hidden layer
  • num_out (int) – the number of units in the output layer
Returns:

the output layer of the DSSM
Return type:

lasagne.layers.Layer
dssm.compute_loss(output, num_samples, num_entries=6, gamma=500.0)[source]

Compute the loss of a dataset, given the output of the DSSM.

Parameters:
  • output (lasagne.layers.Layer) – the output of the DSSM
  • num_samples (int) – the number of samples in the dataset
  • num_entries (int) – the number of compared papers in the DSSM structure
  • gamma (float) – the coefficient applied in the softmax of the similarities
Returns:

the loss of the dataset
Return type:

theano.tensor.TensorType
dssm.iterate_minibatches(inputs, batchsize, shuffle=False)[source]

Produces an batch iterator over the input.

Usage:
>>> for batch in iterate_minibatches(inputs, batchsize, shuffle):
>>>    # to stuff
Parameters:
  • inputs (list) – the input list over which iterate
  • batchsize (int) – the size of each batch
  • shuffle (bool) – if True, inputs is shuffled before iteration