Heatmap

Build heatmaps with annotations, diverging palettes, and tight grids for correlation or matrix data.

import matplotlib.pyplot as plt
import numpy as np

import dartwork_mpl as dm

dm.style.use("scientific")

# Generate data
rng = np.random.default_rng(4)
data = rng.uniform(-1, 1, size=(12, 12))
cov = rng.normal(0, 1, size=(200, 8))
cov[:, 4:] += cov[:, :4] * 0.6  # induce correlation structure
corr = np.corrcoef(cov, rowvar=False)

fig = plt.figure(figsize=(dm.cm2in(18), dm.cm2in(9)), dpi=300)
gs = fig.add_gridspec(
    1,
    2,
    width_ratios=[1, 1],
    left=0.08,
    right=0.94,
    top=0.92,
    bottom=0.12,
    wspace=0.28,
)

# Panel A: Diverging heatmap
ax1 = fig.add_subplot(gs[0, 0])
im1 = ax1.imshow(data, cmap="dm.coolwarm", vmin=-1, vmax=1)
cbar1 = fig.colorbar(im1, ax=ax1, fraction=0.04, pad=0.02)
cbar1.set_label("Value", fontsize=dm.fs(0))
ax1.set_title("Random Field (Diverging)", fontsize=dm.fs(1))
ax1.set_xlabel("X Axis", fontsize=dm.fs(0))
ax1.set_ylabel("Y Axis", fontsize=dm.fs(0))

# Panel B: Correlation matrix
ax2 = fig.add_subplot(gs[0, 1])
im2 = ax2.imshow(corr, cmap="dm.Spectral", vmin=-1, vmax=1)
cbar2 = fig.colorbar(im2, ax=ax2, fraction=0.04, pad=0.02)
cbar2.set_label("Correlation", fontsize=dm.fs(0))
ax2.set_xticks(range(corr.shape[0]))
ax2.set_yticks(range(corr.shape[0]))
ax2.set_xticklabels(
    [f"F{i + 1}" for i in range(corr.shape[0])],
    fontsize=dm.fs(-2),
    rotation=45,
    ha="right",
)
ax2.set_yticklabels(
    [f"F{i + 1}" for i in range(corr.shape[0])], fontsize=dm.fs(-2)
)
ax2.set_title("Feature Correlation Matrix", fontsize=dm.fs(1))

dm.simple_layout(fig, gs=gs)
plt.show()