bebi103.viz.predictive_regression

bebi103.viz.predictive_regression(samples, samples_x, data=None, diff=False, percentiles=[95, 68], color='blue', data_kwargs=None, p=None, **kwargs)

Plot a predictive regression plot from samples.

Parameters

  • samples (Numpy array, shape (n_samples, n_x) or xarray DataArray) – Numpy array containing predictive samples of y-values.

  • sample_x (Numpy array, shape (n_x,)) –

  • data (Numpy array, shape (n, 2) or xarray DataArray) – If not None, the measured data. The first column is the x-data, and the second the y-data. These are plotted as points over the predictive plot.

  • diff (bool, default True) – If True, the predictive y-values minus the median of the predictive y-values are plotted.

  • percentiles (list, default [95, 68]) – Percentiles for making colored envelopes for confidence intervals for the predictive ECDFs. Maximally four can be specified.

  • color (str, default 'blue') – One of [‘green’, ‘blue’, ‘red’, ‘gray’, ‘purple’, ‘orange’]. There are used to make the color scheme of shading of percentiles.

  • data_kwargs (dict, default None) – Any kwargs to be passed to p.circle() when plotting the data points.

  • p (bokeh.plotting.Figure instance, or None (default)) – If None, create a new figure. Otherwise, populate the existing figure p.

  • kwargs – All other kwargs are passed to bokeh.plotting.figure().

Returns

output – Figure populated with glyphs describing range of values for the the samples. The shading goes according to percentiles of samples, with the median plotted as line in the middle.

Return type

Bokeh figure