j_hartling/paper_2025
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Made fig_invariance_cross_species_thresh__appendix.pdf.

Browse Source
This commit is contained in:
j-hartling
2026-04-30 19:34:37 +02:00
parent ca23d42f5d
commit 9c5811d97c
20 changed files with 820 additions and 103 deletions
Download Patch File Download Diff File Expand all files Collapse all files

75
python/fig_features_cross_species.py
View File

@@ -54,9 +54,9 @@ calculate_regression = True
test_regression = True
# SUBSET SETTINGS:
types = np.array([1, -1, 2, -2, 3, -3, 4, -4, 5, -5, 6, -6, 7, -7, 8, -8, 9, -9, 10, -10])
types = np.array([1, -1, 2, -2, 3, -3, 4, -4])
# types = [1, -1, 2, -2, 3, -3, 4, -4, 5, -5, 6, -6, 7, -7, 8, -8, 9, -9, 10, -10]
sigmas = np.array([0.001, 0.002, 0.004, 0.008, 0.016, 0.032])
sigmas = np.array([0.001, 0.002, 0.004, 0.008, 0.016])
# sigmas = [0.001, 0.002, 0.004, 0.008, 0.016, 0.032]
kernels = None
reduce_kernels = any(var is not None for var in [kernels, types, sigmas])
@@ -84,7 +84,7 @@ song_grid_kwargs = dict(
)
# PLOT SETTINGS:
kern_colors = load_colors('../data/feat_colors_all.npz')
kern_colors = load_colors('../data/feat_colors_subset.npz')
fs = dict(
lab_norm=16,
lab_tex=20,
@@ -178,16 +178,44 @@ if test_regression:
test_ax_side,
test_ax_side
]
xlab_test = '$\\rho$'
ylab_test = '$\\text{PDF}_{\\rho}$'
xloc_test = 0.5
yloc_test = 10
ylab_test = '$\\rho$'
yloc_test = 0.5
ylab_test_kwargs = dict(
x=-0.3,
fontsize=fs['lab_norm'],
ha='center',
va='bottom',
)
boxplot_kwargs = dict(
positions=[0, 1],
widths=0.9,
tick_labels=['inter', 'intra'],
zorder=1,
medianprops=dict(
color='k',
lw=1,
),
boxprops=dict(
color='k',
lw=1,
),
)
boxplot_kwargs.update(
capprops=boxplot_kwargs['boxprops'],
whiskerprops=boxplot_kwargs['boxprops'],
)
boxplot_dot_kwargs = dict(
ls='none',
marker='o',
ms=4,
mec='k',
mfc='w',
mew=1.5,
alpha=0.5,
zorder=2,
)
nbins = 10
spec_color = 'darkorchid'
song_color = 'goldenrod'
@@ -385,35 +413,22 @@ for x, y in product(range(n_song), range(n_song)):
if test_regression:
# Add test result subplot:
test_ax = fig.add_subplot(test_ax_bounds)
test_ax.xaxis.set_major_locator(plt.MultipleLocator(xloc_test))
test_ax.set_xlim(-0.6, 1.6)
test_ax.set_ylim(0, 1)
test_ax.yaxis.set_major_locator(plt.MultipleLocator(yloc_test))
xlabel(test_ax, xlab_test, transform=fig.transFigure, **xlab_low_kwargs)
ylabel(test_ax, ylab_test, **ylab_test_kwargs)
# Show boxplots of correlation coefficients:
test_ax.boxplot([spec_regs, song_regs], **boxplot_kwargs)
# Show underlying datapoints:
test_ax.plot(np.zeros(len(spec_regs)), spec_regs, **boxplot_dot_kwargs)
test_ax.plot(np.ones(len(song_regs)), song_regs, **boxplot_dot_kwargs)
# Perform t-test:
test = ttest_ind(spec_regs, song_regs, equal_var=False)
t, p = test.pvalue, test.statistic
print(f'\nT-test result: t={t}, p={p}')
# Calculate histograms:
limits = np.array([min(spec_regs + song_regs), max(spec_regs + song_regs)])
limits += np.array([-1.1, 1.1]) * (limits[1] - limits[0])
edges = np.linspace(*limits, nbins + 1)
centers = edges[:-1] + (edges[1] - edges[0]) / 2
spec_hist, _ = np.histogram(spec_regs, bins=edges, density=True)
song_hist, _ = np.histogram(song_regs, bins=edges, density=True)
# Plot histograms:
bar_kwargs['width'] *= (centers[1] - centers[0])
test_ax.bar(centers, spec_hist, color=spec_color, label='inter-species', **bar_kwargs)
test_ax.bar(centers, song_hist, color=song_color, label='intra-species', **bar_kwargs)
# Indicate means:
test_ax.axvline(np.mean(spec_regs), color=spec_color, **mean_kwargs)
test_ax.axvline(np.mean(song_regs), color=song_color, **mean_kwargs)
# Add legend:
test_ax.legend(**leg_kwargs)
# Posthocs:
test_ax.set_ylim(0, max(spec_hist.max(), song_hist.max()) * 1.05)
test_ax.set_xlim(min(0, max(-1, limits[0])),
min(1, limits[1]))
if save_path is not None:
fig.savefig(save_path)