plotastro 1.1.0__tar.gz → 1.1.0b2__tar.gz

{plotastro-1.1.0 → plotastro-1.1.0b2}/.gitignore

@@ -10,4 +10,6 @@ docs/_build/
 docs/_figures/
 docs/tutorial.ipynb
 .vscode/
+.gitignore
 .DS_Store
+.gitignore

{plotastro-1.1.0 → plotastro-1.1.0b2}/CHANGELOG.md

@@ -1,26 +1,5 @@
 # Changelog
 
-## 1.1.0 — 2026-10-06
-
-The stable release of the 1.1 series: everything in the 1.1.0b1 and
-1.1.0b2 betas below, plus documentation. `pip install plotastro` now
-gives this version. No change in behaviour since 1.1.0b2.
-
-### Added
-- Install instructions for conda-forge
-  (`conda install -c conda-forge plotastro`, Python 3.11 or newer).
-- The API reference is split into one page per topic (styles, colours,
-  CMasher, markers, authors, accessibility). The colours page has a
-  longer guide to colormaps: which map suits which kind of data, lines
-  coloured by a parameter with a colour bar, diverging and cyclic data,
-  and common errors. The tutorial notebook runs the same examples.
-- The docstrings of `lighten`, `darken`, `check_colors`,
-  `current_journal`, `subplots`, `set_size`, `style_cycler`,
-  `show_colors`, `show_markers` and `show_linestyles` now list their
-  parameters and return values.
-- A test checks that every public name is in the API reference and that
-  the constants it shows match the code.
-
 ## 1.1.0b2 — 2026-10-06 (beta)
 
 A pre-release: `pip install plotastro` still gives the stable 1.0.1. To try

{plotastro-1.1.0 → plotastro-1.1.0b2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: plotastro
-Version: 1.1.0
+Version: 1.1.0b2
 Summary: Publication-quality matplotlib styles and helpers for astronomy journals (MNRAS, A&A, ApJ, OJA, PRD, JCAP, Nature Astronomy)
 Project-URL: Homepage, https://github.com/BehnoodBandi/plotastro
 Project-URL: Issues, https://github.com/BehnoodBandi/plotastro/issues
@@ -28,7 +28,6 @@ Description-Content-Type: text/markdown
 # plotastro
 
 [![PyPI](https://img.shields.io/pypi/v/plotastro.svg)](https://pypi.org/project/plotastro/)
-[![conda-forge](https://img.shields.io/conda/vn/conda-forge/plotastro.svg)](https://anaconda.org/conda-forge/plotastro)
 [![Python versions](https://img.shields.io/pypi/pyversions/plotastro.svg)](https://pypi.org/project/plotastro/)
 [![CI](https://github.com/BehnoodBandi/plotastro/actions/workflows/ci.yml/badge.svg)](https://github.com/BehnoodBandi/plotastro/actions/workflows/ci.yml)
 [![Docs](https://readthedocs.org/projects/plotastro/badge/?version=latest)](https://plotastro.readthedocs.io)
@@ -44,7 +43,7 @@ helpers that make the tedious parts (sizing, panel labels, accessibility
 checks, saving) one-liners.
 
 ```bash
-pip install plotastro                      # or: conda install -c conda-forge plotastro
+pip install plotastro
 ```
 
 **Simplest usage — no new API to learn.** Importing plotastro registers the
@@ -275,12 +274,6 @@ map with a fixed number of lines, sample exactly that many:
 `pa.cmasher_colors("rainforest", n=len(models))`. Please cite CMasher if
 you use it (`cmasher.get_bibtex()`).
 
-The [colours page of the documentation](https://plotastro.readthedocs.io/en/latest/colors.html)
-has more: which map suits which kind of data, lines coloured by a
-parameter with a colour bar, diverging and cyclic data, and common
-errors. The [tutorial notebook](examples/tutorial.ipynb) runs the same
-examples.
-
 ### Checking accessibility yourself
 
 ![CVD check](examples/figures/cvd_check.png)

{plotastro-1.1.0 → plotastro-1.1.0b2}/README.md

@@ -1,7 +1,6 @@
 # plotastro
 
 [![PyPI](https://img.shields.io/pypi/v/plotastro.svg)](https://pypi.org/project/plotastro/)
-[![conda-forge](https://img.shields.io/conda/vn/conda-forge/plotastro.svg)](https://anaconda.org/conda-forge/plotastro)
 [![Python versions](https://img.shields.io/pypi/pyversions/plotastro.svg)](https://pypi.org/project/plotastro/)
 [![CI](https://github.com/BehnoodBandi/plotastro/actions/workflows/ci.yml/badge.svg)](https://github.com/BehnoodBandi/plotastro/actions/workflows/ci.yml)
 [![Docs](https://readthedocs.org/projects/plotastro/badge/?version=latest)](https://plotastro.readthedocs.io)
@@ -17,7 +16,7 @@ helpers that make the tedious parts (sizing, panel labels, accessibility
 checks, saving) one-liners.
 
 ```bash
-pip install plotastro                      # or: conda install -c conda-forge plotastro
+pip install plotastro
 ```
 
 **Simplest usage — no new API to learn.** Importing plotastro registers the
@@ -248,12 +247,6 @@ map with a fixed number of lines, sample exactly that many:
 `pa.cmasher_colors("rainforest", n=len(models))`. Please cite CMasher if
 you use it (`cmasher.get_bibtex()`).
 
-The [colours page of the documentation](https://plotastro.readthedocs.io/en/latest/colors.html)
-has more: which map suits which kind of data, lines coloured by a
-parameter with a colour bar, diverging and cyclic data, and common
-errors. The [tutorial notebook](examples/tutorial.ipynb) runs the same
-examples.
-
 ### Checking accessibility yourself
 
 ![CVD check](examples/figures/cvd_check.png)

plotastro-1.1.0b2/docs/api.md

@@ -0,0 +1,94 @@
+# API reference
+
+Everything is available at the top level:
+
+```python
+import plotastro as pa
+```
+
+## Styles and sizing
+
+```{eval-rst}
+.. autofunction:: plotastro.set_style
+.. autofunction:: plotastro.figsize
+.. autofunction:: plotastro.subplots
+.. autofunction:: plotastro.savefig
+.. autofunction:: plotastro.current_journal
+```
+
+`pa.use(...)` is an alias of {func}`plotastro.set_style`.
+
+## Colours
+
+```{eval-rst}
+.. autofunction:: plotastro.lighten
+.. autofunction:: plotastro.darken
+.. autofunction:: plotastro.simulate_cvd
+.. autofunction:: plotastro.check_colors
+.. autofunction:: plotastro.check_figure
+.. autofunction:: plotastro.euclid_colors
+```
+
+### CMasher (optional)
+
+These need the optional `cmasher` package (`pip install cmasher`); see
+{doc}`colors`. `set_style(palette="cmr.<name>", cmap="cmr.<name>")` uses
+them too.
+
+```{eval-rst}
+.. autofunction:: plotastro.cmasher_colors
+.. autofunction:: plotastro.cmasher_cmap
+```
+
+### Palette constants
+
+| name | contents |
+|---|---|
+| `pa.COLORS` | the default 12-colour colour-blind-friendly cycle, by name |
+| `pa.CYCLE` | the same colours as an ordered list |
+| `pa.OKABE_ITO` | Okabe & Ito (2008) 8-colour CVD-safe palette |
+| `pa.PETROFF10` | Petroff (2021) 10-colour CVD-optimised palette |
+| `pa.PETROFF8` | Petroff (2021) 8-colour palette — the Euclid niceplots default |
+| `pa.TOL_VIBRANT` | Paul Tol's *vibrant* 7-colour CVD-safe scheme |
+| `pa.PAIRED` | light/dark pairs: `pa.PAIRED["blue"] -> (light, dark)` |
+
+## Markers, line styles and labels
+
+```{eval-rst}
+.. autofunction:: plotastro.style_cycler
+.. autofunction:: plotastro.label_panels
+```
+
+| name | contents |
+|---|---|
+| `pa.MARKERS` | marker sequence that stays distinguishable at 4 pt |
+| `pa.LINESTYLES` | named dash patterns beyond matplotlib's built-ins |
+
+## Reference charts
+
+```{eval-rst}
+.. autofunction:: plotastro.show_colors
+.. autofunction:: plotastro.show_markers
+.. autofunction:: plotastro.show_linestyles
+```
+
+## Author lists
+
+```{eval-rst}
+.. autofunction:: plotastro.authorlist
+```
+
+The `plotastro-authors` command-line tool wraps this function; run
+`plotastro-authors --help` for its options.
+
+## Journal data and legacy
+
+```{eval-rst}
+.. autofunction:: plotastro.set_size
+```
+
+| name | contents |
+|---|---|
+| `pa.JOURNALS` | per-journal column/full widths (LaTeX points) and metadata |
+| `pa.GOLDEN` | the golden ratio (default figure aspect), ≈ 0.618 |
+| `pa.STYLE_DIR` | path to the bundled `.mplstyle` files |

plotastro-1.1.0b2/docs/colors.md

@@ -0,0 +1,190 @@
+# Colours
+
+## The default palette
+
+```{image} _figures/palette.png
+:alt: The default colour-blind-friendly cycle
+:width: 85%
+```
+
+The default cycle has 12 colours, all accessible by name via
+`plotastro.COLORS` (e.g. `pa.COLORS["blue"]`), or as matplotlib's
+`"C0"`…`"C11"` shorthands:
+
+- **C0–C8** are a colour-blind-safe re-ordering of the
+  [ColorBrewer](https://colorbrewer2.org) *Set1* qualitative palette
+  (popularised by [Thøger Rivera-Thorsen's CBcycle](https://gist.github.com/thriveth/8560036)).
+  Consecutive colours differ in **lightness as well as hue**, so adjacent
+  lines stay distinguishable under the common deficiencies (deuteranopia,
+  protanopia) *and* in greyscale print; the notorious red–green pair is
+  pushed far apart in the cycle (green is C2, red is C7), so plots with a
+  handful of lines never rely on it.
+- **C9–C11** are light companions (from Tableau's *Color Blind 10*): use
+  them for uncertainty bands, reference curves, or de-emphasised data
+  underneath a saturated line of the same hue.
+
+## Matched shades without transparency
+
+Better for print and EPS than `alpha=` (no colour shifts where elements
+overlap):
+
+```python
+ax.plot(x, y, color=pa.COLORS["blue"])
+ax.fill_between(x, lo, hi, color=pa.lighten(pa.COLORS["blue"], 0.7))
+pa.darken(pa.COLORS["orange"], 0.3)     # the other direction
+```
+
+## More palettes
+
+- `pa.OKABE_ITO` — [Okabe & Ito (2008)](https://jfly.uni-koeln.de/color/),
+  *the* classic CVD-safe recommendation for categorical colours in science;
+- `pa.PETROFF10` — [Petroff (2021)](https://arxiv.org/abs/2107.02270), the
+  CVD-optimised 10-colour cycle used across particle physics;
+- `pa.PETROFF8` — Petroff's 8-colour sibling, the default cycle of the
+  Euclid Consortium's [niceplots](https://gitlab.euclid-sgs.uk/ECEB/niceplots)
+  (and of the `euclid` style here);
+- `pa.TOL_VIBRANT` — [Paul Tol's](https://personal.sron.nl/~pault/) *vibrant*
+  qualitative scheme, 7 CVD-safe colours;
+- `pa.PAIRED` — light/dark pairs for data/model or before/after
+  comparisons: `pa.PAIRED["blue"]` → `("#a6cee3", "#1f78b4")`.
+
+Any of them can become the active cycle when you activate a style —
+`pa.set_style("mnras", palette="okabe_ito")` — or pass your own list of
+colours.
+
+## Euclid colour schemes
+
+The `euclid` style (see {doc}`journals`) and these colour schemes are
+adapted from the Euclid Consortium Editorial Board's
+[niceplots](https://gitlab.euclid-sgs.uk/ECEB/niceplots) (Euclid-internal,
+GPL-3.0): the colours are re-expressed here, nothing is copied from it.
+{func}`plotastro.euclid_colors` returns each scheme under its niceplots name:
+
+| scheme | colours |
+|---|---|
+| `"categorical1"` | Petroff (2021) 8 colours — `pa.PETROFF8`; the Euclid default |
+| `"categorical2"` | Okabe & Ito — `pa.OKABE_ITO` |
+| `"categorical3"` | black, then Tol's *vibrant* scheme — `pa.TOL_VIBRANT` |
+| `"sequential"` | `n` colours of increasing brightness from `copper` |
+| `"diverging"` | `n` colours from blue to red from `coolwarm` |
+
+```python
+pa.set_style("euclid", palette="categorical3")                 # by name
+ax.set_prop_cycle(color=pa.euclid_colors("sequential", n=6))   # per axes
+```
+
+## Checking accessibility yourself
+
+```{image} _figures/cvd_check.png
+:alt: The default palette under simulated colour-vision deficiencies
+:width: 85%
+```
+
+Don't take the palette's word for it — simulate it (Machado et al. 2009
+model, no extra dependencies):
+
+```python
+pa.check_colors()                 # any palette under deuteranopia/protanopia/greyscale
+pa.check_colors(pa.PAIRED)        # works on your own colour lists/dicts too
+pa.check_figure(fig)              # simulate a whole rendered figure — the
+                                  # final check before submission
+pa.simulate_cvd("#e41a1c", "deuteranopia")   # the raw transform
+```
+
+If two lines merge in any panel, add markers or dash patterns (see
+{doc}`markers`), or pick colours further apart in the cycle. MNRAS
+recommends [Color Oracle](https://colororacle.org) and ColorBrewer for
+exactly this; with plotastro it's built in.
+
+## Colormaps
+
+The styles default to `viridis` (perceptually uniform, CVD-safe). Good
+picks: `viridis`/`magma`/`cividis` for sequential data, `RdBu_r` or
+`coolwarm` for diverging data (red–*blue*, not red–green). Avoid
+`jet`/`rainbow`. To change the default for every `imshow`, `pcolormesh`,
+`scatter`, … pass `cmap=` when you activate a style:
+
+```python
+pa.set_style("mnras", cmap="cividis")
+```
+
+For many more maps, plotastro can use CMasher (next section); see also
+[cmocean](https://matplotlib.org/cmocean/).
+
+## CMasher colours and colormaps (optional)
+
+```{image} _figures/cmasher.png
+:alt: A CMasher colour cycle for a family of lines, and a CMasher colormap on an image
+:width: 100%
+```
+
+[CMasher](https://cmasher.readthedocs.io) (van der Velden 2020,
+[JOSS 5, 2004](https://doi.org/10.21105/joss.02004)) is a collection of
+scientific colormaps: sequential, diverging and cyclic, all designed to be
+perceptually uniform, and most of them colour-vision-deficiency friendly.
+plotastro can use them for discrete colours and for colormaps. CMasher is
+**not** a dependency of plotastro: install it only if you want it,
+
+```bash
+pip install cmasher                # or: pip install "plotastro[cmasher]"
+```
+
+and plotastro imports it only when you ask for a CMasher colour. Every
+other feature works the same without it. CMasher names start with
+`cmr.` (`"cmr.rainforest"`, `"cmr.iceburn"`, …), which is also how its
+maps are registered with matplotlib; append `_r` for the reversed map.
+Browse them all in the
+[CMasher colormap overview](https://cmasher.readthedocs.io), or list them
+with `cmasher.get_cmap_list()`.
+
+### Discrete colours
+
+Pass a `"cmr."` name as the palette to get a colour cycle of 8 colours
+sampled from that map, or use {func}`plotastro.cmasher_colors` for a
+different number:
+
+```python
+pa.set_style("mnras", palette="cmr.rainforest")              # 8-colour cycle
+ax.set_prop_cycle(color=pa.cmasher_colors("torch", n=5))     # 5, this axes only
+colors = pa.cmasher_colors("ocean", n=4, cmap_range=(0.2, 0.8))
+```
+
+The colours are hex strings, equally spaced over `cmap_range`. Its default,
+`(0.15, 0.85)`, follows CMasher's own advice: most of its sequential maps
+run from black to white, and those ends disappear against the axes or the
+page. Sequential maps work best for lines. CMasher recommends:
+
+- **for lines that must be easy to tell apart:** maps with a large
+  perceptual range, such as `apple`, `chroma`, `neon`, `rainforest` and
+  `torch`;
+- **for lines that are steps of one quantity** (redshifts, masses, …): a
+  single-hue map, such as `flamingo`, `freeze`, `gothic`, `jungle` and
+  `ocean`.
+
+These cycles are ordered from dark to light, and a matplotlib cycle starts
+at the first colour. So for a fixed number of lines, sample exactly that
+many, `pa.cmasher_colors("rainforest", n=len(models))`, and they will span
+the whole map. Check the result with `pa.check_colors(...)` as for any other
+palette.
+
+### Colormaps
+
+Make a CMasher map the default colormap with `set_style(cmap=...)`, or get
+the colormap itself with {func}`plotastro.cmasher_cmap`. It can also cut
+the map to part of its range, or split it into a few discrete levels:
+
+```python
+pa.set_style("mnras", cmap="cmr.ocean")            # default for imshow etc.
+
+ax.imshow(img, cmap=pa.cmasher_cmap("rainforest"))
+ax.pcolormesh(x, y, z, cmap=pa.cmasher_cmap("ocean", cmap_range=(0.15, 0.85)))
+ax.contourf(x, y, z, levels=6, cmap=pa.cmasher_cmap("iceburn", n=6))
+```
+
+Once CMasher has been imported (by plotastro or by `import cmasher`), every
+matplotlib function also accepts its maps by name: `cmap="cmr.iceburn"`.
+Some of CMasher's diverging maps (`iceburn`, `redshift`, `seaweed`,
+`watermelon`, `wildfire`) have a **black** centre instead of a white one.
+
+If you use CMasher in a paper, please cite it; `cmasher.get_bibtex()`
+prints the reference.

{plotastro-1.1.0 → plotastro-1.1.0b2}/docs/index.md

@@ -10,7 +10,7 @@ helpers that make the tedious parts (sizing, panel labels, accessibility
 checks, author lists, saving) one-liners.
 
 ```bash
-pip install plotastro                      # or: conda install -c conda-forge plotastro
+pip install plotastro
 ```
 
 **Simplest usage — no new API to learn.** Importing plotastro registers the
@@ -72,7 +72,7 @@ deliberately matches the Euclid Consortium's own niceplots look instead
 - {doc}`quickstart` — the five-minute version
 - {doc}`tutorial` — the full hands-on notebook, rendered
 - {doc}`journals` — supported journals and their figure widths
-- {doc}`colors` — the palettes, colour-blindness checking and CMasher colormaps
+- {doc}`colors` — the palettes and colour-blindness checking
 - {doc}`markers` — markers, line styles, cyclers and panel labels
 - {doc}`authors` — LaTeX author lists from your collaboration's CSV
 - {doc}`api` — every public function

{plotastro-1.1.0 → plotastro-1.1.0b2}/docs/installation.md

@@ -6,11 +6,11 @@
 pip install plotastro
 ```
 
-Betas (pre-releases) are opt-in: plain `pip install` gives you the latest
-stable version. When a beta newer than that is out, install it with:
+To try the latest **beta** (a pre-release; plain `pip install` keeps giving
+you the stable version):
 
 ```bash
-pip install --pre plotastro
+pip install --pre plotastro                # or: pip install "plotastro==1.1.0b2"
 pip install --pre "plotastro[cmasher]"     # the beta with CMasher
 ```
 
@@ -31,16 +31,6 @@ pip install "plotastro[cmasher]"     # or simply: pip install cmasher
 plotastro imports CMasher only when you ask for a CMasher colour, so it is
 never needed otherwise. (Installing with conda? `conda install -c conda-forge cmasher`.)
 
-## From conda-forge
-
-```bash
-conda install -c conda-forge plotastro             # or: mamba install -c conda-forge plotastro
-conda install -c conda-forge plotastro cmasher     # with CMasher
-```
-
-The conda-forge package needs Python 3.11 or newer (conda-forge's current
-minimum); on older Pythons, install with pip instead.
-
 ## From a clone
 
 ```bash

plotastro-1.1.0b2/examples/make_reference_figures.py

@@ -0,0 +1,117 @@
+"""Generate the reference figures embedded in README.md.
+
+Run from the repository root:  python examples/make_reference_figures.py
+"""
+
+import sys
+from pathlib import Path
+
+import matplotlib
+
+matplotlib.use("Agg")
+import matplotlib.pyplot as plt
+import numpy as np
+
+try:
+    import plotastro as pa
+except ImportError:  # running from a source checkout without installing
+    sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "src"))
+    import plotastro as pa
+
+FIGDIR = Path(__file__).resolve().parent / "figures"
+FIGDIR.mkdir(exist_ok=True)
+rng = np.random.default_rng(42)
+
+
+def save(fig, name):
+    fig.savefig(FIGDIR / f"{name}.png", dpi=300)
+    plt.close(fig)
+    print(f"wrote figures/{name}.png")
+
+
+# ---------------------------------------------------------------- column plot
+pa.set_style("mnras")
+
+x = np.linspace(0.5, 10, 18)
+truth = 2.0 * x ** -0.7
+y = truth * rng.normal(1, 0.08, x.size)
+yerr = 0.08 * truth
+xf = np.linspace(0.4, 11, 200)
+
+fig, ax = pa.subplots()
+ax.errorbar(x, y, yerr=yerr, fmt="o", color=pa.COLORS["blue"],
+            label="mock data", zorder=3)
+ax.plot(xf, 2.0 * xf ** -0.7, color=pa.COLORS["red"], label=r"$2\,x^{-0.7}$")
+ax.fill_between(xf, 1.8 * xf ** -0.7, 2.2 * xf ** -0.7,
+                color=pa.lighten(pa.COLORS["red"], 0.75), zorder=0,
+                label=r"$1\sigma$ band")
+ax.set_xscale("log")
+ax.set_yscale("log")
+ax.set_xlabel(r"$r\ \mathrm{[Mpc]}$")
+ax.set_ylabel(r"$\xi(r)$")
+ax.legend()
+save(fig, "example_column")
+
+# ------------------------------------------------------------ full-width plot
+k = np.logspace(-3, 1, 300)
+fig, axes = pa.subplots(1, 2, width="full", aspect=0.75)
+for i, z in enumerate([0, 0.5, 1, 2]):
+    pk = 1e4 * k / (1 + (k / 0.02) ** 2.2) / (1 + z) ** 1.5
+    axes[0].loglog(k, pk, label=f"$z={z}$")
+    axes[1].semilogx(k, pk / (1e4 * k / (1 + (k / 0.02) ** 2.2)),
+                     label=f"$z={z}$")
+axes[0].set_xlabel(r"$k\ [h\,\mathrm{Mpc}^{-1}]$")
+axes[0].set_ylabel(r"$P(k)\ [h^{-3}\,\mathrm{Mpc}^{3}]$")
+axes[1].set_xlabel(r"$k\ [h\,\mathrm{Mpc}^{-1}]$")
+axes[1].set_ylabel(r"$P(k)/P(k, z=0)$")
+axes[0].legend()
+pa.label_panels(axes)
+save(fig, "example_full")
+
+# ------------------------------------------------- palette / marker reference
+save(pa.show_colors(), "palette")
+save(pa.show_colors(pa.OKABE_ITO, title="Okabe & Ito (2008) — plotastro.OKABE_ITO"),
+     "palette_okabe_ito")
+save(pa.show_markers(), "markers")
+save(pa.show_linestyles(), "linestyles")
+
+# ------------------------------------------------------- redundant encoding
+x = np.linspace(0, 3, 60)
+fig, ax = pa.subplots()
+ax.set_prop_cycle(pa.style_cycler(markers=True, linestyles=True))
+for n in range(4):
+    ax.plot(x, x ** (0.5 + 0.4 * n), markevery=7, label=f"model {n + 1}")
+ax.set_xlabel("$x$")
+ax.set_ylabel("$y$")
+ax.legend()
+save(fig, "redundant_encoding")
+
+# --------------------------------------------------------------- CVD check
+save(pa.check_colors(), "cvd_check")
+
+# ------------------------------------------- CMasher colours (optional extra)
+try:
+    import cmasher  # noqa: F401  (only to check it is installed)
+except ImportError:
+    print("cmasher not installed: skipping figures/cmasher.png")
+else:
+    pa.set_style("mnras", palette="cmr.rainforest", cmap="cmr.ocean")
+    fig, axes = pa.subplots(1, 2, width="full", aspect=0.75)
+    x = np.linspace(0, 1, 200)
+    for n in range(8):                       # the 8-colour CMasher cycle
+        axes[0].plot(x, x ** (0.4 + 0.3 * n), label=f"$n={n + 1}$")
+    axes[0].set_xlabel("$x$")
+    axes[0].set_ylabel("$x^{\\alpha_n}$")
+    axes[0].set_title("palette='cmr.rainforest'", family="monospace", fontsize=8)
+    axes[0].legend(ncol=2, fontsize=6)
+    yy, xx = np.mgrid[-3:3:200j, -3:3:200j]
+    field = (np.exp(-((xx - 0.8) ** 2 + yy ** 2)) + 0.6 * np.exp(
+        -((xx + 1.2) ** 2 + (yy - 1) ** 2) / 0.5))
+    im = axes[1].imshow(field, origin="lower", extent=(-3, 3, -3, 3))
+    axes[1].grid(False)
+    axes[1].set_xlabel("$x$")
+    axes[1].set_ylabel("$y$")
+    axes[1].set_title("cmap='cmr.ocean'", family="monospace", fontsize=8)
+    fig.colorbar(im, ax=axes[1], label="density")
+    pa.label_panels(axes, loc="lower right")[1].set_color("white")
+    save(fig, "cmasher")