pyobsmod.plots.scatter_plot_sns

pyobsmod.plots.scatter_plot_sns(obs: ndarray, mod: ndarray, which_stats: Sequence[str] | dict[str, Any] | None = None, names: Sequence[str] | None = None, fmt: str | Sequence[str] = '.2f', idline_kws: dict[str, Any] | None = None, textbox_kws: dict[str, Any] | None = None, **kwargs) → JointGrid

Scatter plot sns of observed data against modelled data.

Add an identity line and display selected statistics with seaborn’s jointplot.

Internally this function create a Dataset object and passes the arguments to the Dataset.scatter_plot_sns method.

Parameters:

• obs (numpy.ndarray) – 1d numpy array that contains the data that was observed / measured.

• mod (numpy.ndarray) – 1d numpy array that contains the data that was predicted / modelled.

• which_stats (Sequence[str] | dict[str, Any] | None) – Sequence of the statistics parameters to compute or alternatively a dictionary with the statistics parameters as keys and the arguments that are passed to the method as values.

• names (Sequence[str] | None) – Sequence of the names of the statistics parameters. If None, the names are taken from the keys of the which_stats dictionary.

• fmt (str | Sequence[str]) – Format string for the statistics. Can be a single string that is used for all statistics or a sequence of strings that is used for

• idline_kws (dict[str, Any] | None) – Dictionary that is passed to matplotlib.pyplot.axline.

• textbox_kws (dict[str, Any] | None) – Dictionary that is passed to matplotlib.offsetbox.AnchoredText.

• **kwargs – Additional arguments that are passed to seaborn.jointplot.

Returns:

grid – An instance of a seaborn JointGrid.

Return type:

seaborn.JointGrid

Notes

Note that in contrast to plots.scatter_plot and plots.time_series_plot this method does NOT take a matplotlib.pyplot.Axes object as an argument, since the underlying function creates a figure and several axes objects itself.

In practice this means simply, that this plot can NOT be used in subplots (without major work-arounds).

Examples

import matplotlib.pyplot as plt
import numpy as np
import pyobsmod.plots as pp

obs = np.sin(np.arange(100)) + np.random.normal(size=100)
mod = np.sin(np.arange(100)) + np.random.normal(size=100)
g = pp.scatter_plot_sns(obs, mod, ['bias', 'rmse', 'nrmse', 'r2'])
plt.show()