morphoclass.deepwalk module

Portal to DeepWalk methods from morphoclass.

morphoclass.deepwalk.check_installed() → bool

Check whether the DeepWalk package is installed.

Returns

Whether the DeepWalk package is installed

Return type

bool

morphoclass.deepwalk.get_embedding(tree, undirected, number_walks, walk_length, max_memory_data_size, seed, representation_size, window_size, workers)

Learn representations for tree vertices in graphs using DeepWalk.

Parameters

  • tree (tmd.Tree.Tree.Tree) – TMD tree of the neuron.

  • undirected (bool) – Treat graph as undirected.

  • number_walks (int) – Number of random walks to start at each node.

  • walk_length (int) – Length of the random walk started at each node.

  • max_memory_data_size (int) – Size to start dumping walks to disk, instead of keeping them in memory.

  • seed (int) – Seed for random walk generator.

  • representation_size (int) – Number of latent dimensions to learn for each node.

  • window_size (int) – Window size of skipgram model.

  • workers (int) – Number of parallel processes.

Returns

Embedded vectors.

Return type

numpy.ndarray

References

See file: - https://github.com/phanein/deepwalk/blob/ 6e6dff245e4692e9bea47e9017c1034e51afbf29/deepwalk/__main__.py in case max_memory_data_size is ever crossed.

Raises

NotImplementedError – The part was not implemented since we don’t have big graphs size that are not able to fit into the memory.

morphoclass.deepwalk.how_to_install_msg() → str

Get installation instructions for DeepWalk.

Returns

The instructions on how to install DeepWalk.

Return type

str

morphoclass.deepwalk.warn_if_not_installed() → None

Issue a UserWarning if deepwalk is not installed.