simba.pl.pbg_metrics

simba.pl.pbg_metrics(metrics=['mrr'], path_emb=None, fig_size=(5, 3), fig_ncol=1, save_fig=None, fig_path=None, fig_name='pbg_metrics.pdf', pad=1.08, w_pad=None, h_pad=None, **kwargs)[source]

Plot PBG training metrics

Parameters:

  • metrics (list, optional (default: [‘mrr])) –

    Evalulation metrics for PBG training. Possible metrics:

    • ’pos_rank’ : the average of the ranks of all positives (lower is better, best is 1).

    • ’mrr’ : the average of the reciprocal of the ranks of all positives (higher is better, best is 1).

    • ’r1’the fraction of positives that rank better than

      all their negatives, i.e., have a rank of 1 (higher is better, best is 1).

    • ’r10’the fraction of positives that rank in the top 10

      among their negatives (higher is better, best is 1).

    • ’r50’the fraction of positives that rank in the top 50

      among their negatives (higher is better, best is 1).

    • ’auc’ : Area Under the Curve (AUC)

  • path_emb (str, optional (default: None)) – Path to directory for pbg embedding model. If None, .settings.pbg_params[‘checkpoint_path’] will be used.

  • pad (float, optional (default: 1.08)) – Padding between the figure edge and the edges of subplots, as a fraction of the font size.

  • h_pad (float, optional (default: None)) – Padding (height/width) between edges of adjacent subplots, as a fraction of the font size. Defaults to pad.

  • w_pad (float, optional (default: None)) – Padding (height/width) between edges of adjacent subplots, as a fraction of the font size. Defaults to pad.

  • fig_size (tuple, optional (default: (5, 3))) – figure size.

  • fig_ncol (int, optional (default: 1)) – the number of columns of the figure panel

  • save_fig (bool, optional (default: False)) – if True,save the figure.

  • fig_path (str, optional (default: None)) – If save_fig is True, specify figure path.

  • fig_name (str, optional (default: ‘plot_umap.pdf’)) – if save_fig is True, specify figure name.

Return type:

None