"""
The :mod:`chemotools.plotting._feature_selection` module implements the FeatureSelectionPlot class.
"""
from typing import Any, Optional
import numpy as np
from matplotlib.axes import Axes
from matplotlib.figure import Figure
from chemotools.plotting._spectra import SpectraPlot
from chemotools.plotting._utils import validate_data
[docs]
class FeatureSelectionPlot(SpectraPlot):
"""Plot class for visualizing feature selection on spectral data.
This class extends SpectraPlot to highlight excluded features using
colored vertical spans.
Parameters
----------
x : np.ndarray
X-axis data (e.g., wavelengths, wavenumbers).
y : np.ndarray
Y-axis data (e.g., spectra intensities).
support : np.ndarray
Boolean mask indicating selected features (True means selected).
Must have same length as x.
selection_color : str, optional
Color to use for highlighting excluded features. Default is "red".
selection_alpha : float, optional
Transparency of the selection highlight. Default is 0.2.
**kwargs : Any
Additional arguments passed to SpectraPlot.__init__.
"""
def __init__(
self,
x: np.ndarray,
y: np.ndarray,
support: np.ndarray,
*,
selection_color: str = "red",
selection_alpha: float = 0.2,
**kwargs: Any,
):
super().__init__(x, y, **kwargs)
self.support = validate_data(
support, name="support", ensure_2d=False, numeric=False
).astype(bool)
if len(self.support) != len(self.x):
raise ValueError(
f"Support mask length ({len(self.support)}) must match x length ({len(self.x)})"
)
self.selection_color = selection_color
self.selection_alpha = selection_alpha
[docs]
def show(
self,
figsize=(12, 4),
title=None,
xlabel="X-axis",
ylabel="Y-axis",
xlim=None,
ylim=None,
**kwargs,
) -> Figure:
"""Show the spectra plot with given figure size and labels. The
excluded features are highlighted with red bars.
Parameters
----------
figsize : tuple, optional
Figure size as (width, height) in inches. Default is (12, 4).
title : str, optional
Title for the plot. If None, a default title is generated.
xlabel : str, optional
X-axis label. Default is "X-axis".
ylabel : str, optional
Y-axis label. Default is "Y-axis".
xlim : tuple, optional
X-axis limits as (xmin, xmax). Default is None (auto).
ylim : tuple, optional
Y-axis limits as (ymin, ymax). Default is None (auto).
**kwargs : Any
Additional keyword arguments passed to the plot function.
Returns
-------
Figure
The matplotlib Figure object containing the plot.
"""
return super().show(
figsize=figsize,
title=title,
xlabel=xlabel,
ylabel=ylabel,
xlim=xlim,
ylim=ylim,
**kwargs,
)
[docs]
def render(
self,
ax: Optional[Axes] = None,
*,
xlabel: Optional[str] = None,
ylabel: Optional[str] = None,
xlim: Optional[tuple[float, float]] = None,
ylim: Optional[tuple[float, float]] = None,
**kwargs: Any,
) -> tuple[Figure, Axes]:
"""Render the plot on the given axes or create new ones.
Parameters
----------
ax : Axes, optional
Matplotlib axes to plot on. If None, creates new figure and axes.
xlabel : str, optional
X-axis label. Default is "X-axis".
ylabel : str, optional
Y-axis label. Default is "Y-axis".
xlim : tuple[float, float], optional
X-axis limits as (xmin, xmax). When set without ylim, the y-axis
automatically scales to fit the data within the x-range.
ylim : tuple[float, float], optional
Y-axis limits as (ymin, ymax). When provided, disables automatic y-scaling.
**kwargs : Any
Additional keyword arguments passed to the plot function.
Returns
-------
fig : Figure
The matplotlib Figure object.
ax : Axes
The matplotlib Axes object with the rendered plot.
"""
fig, ax = super().render(
ax=ax,
xlabel=xlabel,
ylabel=ylabel,
xlim=xlim,
ylim=ylim,
**kwargs,
)
return fig, ax
def _render_plot(self, ax: Axes, **kwargs: Any) -> None:
"""Render the plot with feature selection highlights.
Parameters
----------
ax : Axes
Matplotlib axes to plot on.
**kwargs : Any
Additional keyword arguments passed to the plot function.
"""
# 1. Render the standard spectrum plot
super()._render_plot(ax, **kwargs)
# 2. Overlay the selection regions (highlight excluded features)
regions = self._get_continuous_regions(~self.support)
# Add label only once for the legend
label_added = False
for start_idx, end_idx in regions:
# Get x coordinates for the span
# We extend slightly to cover the full point width if needed,
# but for now using exact point coordinates is standard
x_start = self.x[start_idx]
x_end = self.x[end_idx]
# Handle case where start > end (e.g. wavenumbers in descending order)
if x_start > x_end:
x_start, x_end = x_end, x_start
label = "Excluded Features" if not label_added else None
ax.axvspan(
x_start,
x_end,
color=self.selection_color,
alpha=self.selection_alpha,
label=label,
zorder=-1, # Put behind spectra
)
label_added = True
def _get_continuous_regions(self, mask: np.ndarray) -> list[tuple[int, int]]:
"""Convert boolean mask to list of (start, end) indices."""
# Pad with False to detect edges
padded = np.concatenate(([False], mask, [False]))
# Find where values change
# diff = 1 means False -> True (Start)
# diff = -1 means True -> False (End)
changes = np.diff(padded.astype(int))
starts = np.where(changes == 1)[0]
ends = np.where(changes == -1)[0] - 1
return list(zip(starts, ends))
