WashU-Astroparticle-Lab / straxion / 26244110227

66%

push

github

Add optional log color scale to plot_channels (#87)

* Add optional log color scale to plot_channels

Introduce a ``log_scale`` flag on ``plot_channels`` that switches the
colormap normalization to ``LogNorm``. Non-positive values, which have
no representation on a log scale, are masked and rendered in a
configurable ``bad_color`` (default ``"lightgrey"``) so the channel
location remains visible. ``vmin``/``vmax`` are honored when provided
and otherwise derived from the positive values.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* Use NaN instead of masked array for log_scale non-positives

Avoids matplotlib UserWarnings ("converting a masked element to nan")
emitted on every scatter call. NaN is treated as 'bad' by the colormap
identically to a masked value, so the rendered output is unchanged.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* Render log_scale non-positive channels with a dedicated scatter

cmap.set_bad does not actually paint NaN entries in a PathCollection -
matplotlib scatter drops them, so the previous implementation left
non-positive channels invisible. Draw a separate scatter for the
non-positive subset with an explicit bad_color facecolor, and draw the
positive subset under LogNorm. Tests now assert on the actually
rendered facecolors and the number of plotted points per collection.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

13 of 13 new or added lines in 1 file covered. (100.0%)

1308 of 1952 relevant lines covered (67.01%)

0.67 hits per line