drippy.univariate

Plotting functions for univariate models (y = c + e).

Attributes

_ORDERED_VALUES

Functions

`run_sequence_plot`(→ tuple[matplotlib.figure.Figure, ...)	Creates a run sequence plot of y vs index or continuous index t.
`lag_plot`(→ tuple[matplotlib.figure.Figure, ...)	Creates a lag plot of the data.
`histogram`(→ tuple[matplotlib.figure.Figure, ...)	Creates a histogram of the data.
`normal_probability_plot`(...)	Creates a normal probability plot.
`four_plot`(→ tuple[matplotlib.figure.Figure, numpy.ndarray])	Creates a 4-plot (run sequence, lag, histogram, normal probability).
`ppcc_plot`(, n_rough, n_fine, numpy.ndarray])	Creates a PPCC plot (rough + fine) for distribution shape estimation.
`weibull_plot`(→ tuple[matplotlib.figure.Figure, ...)	Creates a Weibull probability plot.
`probability_plot`(→ tuple[matplotlib.figure.Figure, ...)	Creates a probability plot for a specified distribution.
`box_cox_normality_plot`(...)	Creates a Box-Cox normality plot (2x2 grid).
`bootstrap_plot`(→ tuple[matplotlib.figure.Figure, ...)	Creates a bootstrap distribution plot.
`box_cox_linearity_plot`(, n, matplotlib.axes.Axes])	Creates a Box-Cox linearity plot.

Module Contents

drippy.univariate._ORDERED_VALUES = 'Ordered Values'[source]

drippy.univariate.run_sequence_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: matplotlib.axes.Axes | None = None) → tuple[matplotlib.figure.Figure, matplotlib.axes.Axes][source]

Creates a run sequence plot of y vs index or continuous index t.

Parameters:

data – EDAData container. Uses t as x-axis if present, else index.
fig – Matplotlib figure. If None, creates new figure.
ax – Matplotlib axes. If None, creates new axes.

Returns:

The figure and axes containing the plot.

drippy.univariate.lag_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: matplotlib.axes.Axes | None = None, lag: int = 1) → tuple[matplotlib.figure.Figure, matplotlib.axes.Axes][source]

Creates a lag plot of the data.

Parameters:

data – EDAData container. Requires y.
fig – Matplotlib figure. If None, creates new figure.
ax – Matplotlib axes. If None, creates new axes.
lag – Number of lags. Must be positive and less than len(y).

Returns:

The figure and axes containing the plot.

drippy.univariate.histogram(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: matplotlib.axes.Axes | None = None, bins: int | str = 'auto') → tuple[matplotlib.figure.Figure, matplotlib.axes.Axes][source]

Creates a histogram of the data.

Parameters:

data – EDAData container. Requires y.
fig – Matplotlib figure. If None, creates new figure.
ax – Matplotlib axes. If None, creates new axes.
bins – Number of bins or bin strategy.

Returns:

The figure and axes containing the plot.

drippy.univariate.normal_probability_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: matplotlib.axes.Axes | None = None, *, return_rsquared: bool = False) → tuple[matplotlib.figure.Figure, matplotlib.axes.Axes, float | None][source]

Creates a normal probability plot.

Parameters:

data – EDAData container. Requires y.
fig – Matplotlib figure. If None, creates new figure.
ax – Matplotlib axes. If None, creates new axes.
return_rsquared – If True, includes R-squared as the third element.

Returns:

(fig, ax, r_squared) where r_squared is None if return_rsquared is False.

drippy.univariate.four_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, axes: numpy.ndarray | None = None) → tuple[matplotlib.figure.Figure, numpy.ndarray][source]

Creates a 4-plot (run sequence, lag, histogram, normal probability).

Parameters:

data – EDAData container. Requires y.
fig – Matplotlib figure. If None, creates new figure.
axes – ndarray of Axes with shape (2, 2). If None, creates new axes.

Returns:

(fig, axes_flat) where axes_flat has shape (4,).

drippy.univariate.ppcc_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: numpy.ndarray | None = None, rough_range: tuple[float, float] = (-2, 2), n_rough: int = 50, n_fine: int = 100) → tuple[matplotlib.figure.Figure, numpy.ndarray][source]

Creates a PPCC plot (rough + fine) for distribution shape estimation.

Parameters:

data – EDAData container. Requires y.
fig – Matplotlib figure. If None, creates new figure.
ax – ndarray of Axes with shape (2,). If None, creates new axes.
rough_range – (min, max) range for rough search.
n_rough – Points in rough plot.
n_fine – Points in fine plot.

Returns:

(fig, axes) where axes has shape (2,).

drippy.univariate.weibull_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: matplotlib.axes.Axes | None = None) → tuple[matplotlib.figure.Figure, matplotlib.axes.Axes][source]

Creates a Weibull probability plot.

Parameters:

data – EDAData container. Requires y > 0.
fig – Matplotlib figure. If None, creates new figure.
ax – Matplotlib axes. If None, creates new axes.

Returns:

The figure and axes containing the plot.

drippy.univariate.probability_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: matplotlib.axes.Axes | None = None, distribution: str = 'norm') → tuple[matplotlib.figure.Figure, matplotlib.axes.Axes][source]

Creates a probability plot for a specified distribution.

Parameters:

data – EDAData container. Requires y.
fig – Matplotlib figure. If None, creates new figure.
ax – Matplotlib axes. If None, creates new axes.
distribution – scipy.stats distribution name.

Returns:

The figure and axes containing the plot.

drippy.univariate.box_cox_normality_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, axes: numpy.ndarray | None = None) → tuple[matplotlib.figure.Figure, numpy.ndarray][source]

Creates a Box-Cox normality plot (2x2 grid).

Shows: original histogram, Box-Cox normality curve, transformed histogram, normal probability plot of transformed data.

Parameters:

data – EDAData container. Requires y > 0.
fig – Matplotlib figure. If None, creates new figure.
axes – ndarray of Axes with shape (2, 2). If None, creates new.

Returns:

(fig, axes) where axes has shape (2, 2).

drippy.univariate.bootstrap_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: matplotlib.axes.Axes | None = None, statistic: collections.abc.Callable = np.mean, n_bootstrap: int = 10000) → tuple[matplotlib.figure.Figure, matplotlib.axes.Axes][source]

Creates a bootstrap distribution plot.

Parameters:

data – EDAData container. Requires y.
fig – Matplotlib figure. If None, creates new figure.
ax – Matplotlib axes. If None, creates new axes.
statistic – Callable to bootstrap. Must be callable.
n_bootstrap – Number of bootstrap resamples.

Returns:

The figure and axes containing the plot.

drippy.univariate.box_cox_linearity_plot(data: drippy.data.EDAData, fig: matplotlib.figure.Figure | None = None, ax: matplotlib.axes.Axes | None = None, lmbda_range: tuple[float, float] = (-2, 2), n: int = 100) → tuple[matplotlib.figure.Figure, matplotlib.axes.Axes][source]

Creates a Box-Cox linearity plot.

Plots abs(corr(Y, X^λ)) across a range of λ values to find the power transformation of X that maximises linearity with Y.

Parameters:

data – EDAData container. Requires x > 0.
fig – Matplotlib figure. If None, creates new figure.
ax – Matplotlib axes. If None, creates new axes.
lmbda_range – (min, max) range of λ to evaluate.
n – Number of λ values to evaluate.

Returns:

The figure and axes containing the plot.