morphoclass.console.stats_plotter module

Default data preparation script shared across different datasets.

class morphoclass.console.stats_plotter.DatasetStatsPlotter(dataset_csv_path: pathlib.Path, report_path: pathlib.Path, morph_name_column: str = 'morph_name', morph_path_column: str = 'morph_path', mtype_column: str = 'mtype')

Bases: object

Plotter for morphology dataset statistics.

Parameters

  • dataset_csv_path – The path for the morphology dataset CSV file.

  • report_path – The output path for the HTML report.

property df_stats

Get a data frame with dataset statistics.

The data frame contains the following columns:

  • path: the neuron morphology file path

  • layer: the m-type layer; e.g. the m-type L5_UPC corresponds to layer L5

  • {axon,apical,basal,neurites}-size: the node count in the respective neurite graph

  • {axon,apical,basal,neurites}-pd-size: the point count in the respective persistence diagram.

Returns

pd.DataFrame

Return type

The data frame with dataset statistics

static load_neuron(path: str) → tmd.Neuron.Neuron.Neuron

Load a TMD neuron given a file path.

make_data_count_plots() → None

Make data count plots and save them to disk.

make_neuron_preview_plots() → None

Make neuron preview plots and save them to disk.

For each m-type one sample is selected randomly and plotted. The plot contains two renders of the 3D morphology, one in the front view, and one in the side view, as well as a combined plot of a persistence diagram overlaid on top of a persistence image.

make_node_count_plots(neurite: str) → None

Make neurite graph and PD count plots and save them to disk.

property neurons

Get a map from a neuron morphology file path to the loaded neuron.

run() → None

Run all plotting routines and save all plots to disk.

save_report(report_title: str = 'Dataset Plots') → None

Save the HTML report to disk.

savefig(fig: matplotlib.figure.Figure, path: pathlib.Path) → pathlib.Path

Save a figure and return the input path.