Features¶
The pygsp.features
module implements different feature extraction
techniques based on pygsp.graphs
and pygsp.filters
.
- pygsp.features.compute_avg_adj_deg(G)[source]¶
Compute the average adjacency degree for each node.
The average adjacency degree is the average of the degrees of a node and its neighbors.
- Parameters
- G: Graph
Graph on which the statistic is extracted
- pygsp.features.compute_norm_tig(g, **kwargs)[source]¶
Compute the \(\ell_2\) norm of the Tig. See
compute_tig()
.- Parameters
- g: Filter
The filter or filter bank.
- kwargs: dict
Additional parameters to be passed to the
pygsp.filters.Filter.filter()
method.
- pygsp.features.compute_spectrogram(G, atom=None, M=100, **kwargs)[source]¶
Compute the norm of the Tig for all nodes with a kernel shifted along the spectral axis.
- Parameters
- GGraph
Graph on which to compute the spectrogram.
- atomfunc
Kernel to use in the spectrogram (default = exp(-M*(x/lmax)²)).
- Mint (optional)
Number of samples on the spectral scale. (default = 100)
- kwargs: dict
Additional parameters to be passed to the
pygsp.filters.Filter.filter()
method.
- pygsp.features.compute_tig(g, **kwargs)[source]¶
Compute the Tig for a given filter or filter bank.
\[T_ig(n) = g(L)_{i, n}\]- Parameters
- g: Filter
One of
pygsp.filters
.- kwargs: dict
Additional parameters to be passed to the
pygsp.filters.Filter.filter()
method.