PLT Module (`stx.plt`)

Publication-ready figures powered by figrecipe.

Note

stx.plt wraps matplotlib via the figrecipe package, adding mm-based layout, data-tracking axes, auto CSV export, and reproducible YAML recipes. The standalone package can also be used directly: pip install figrecipe.

Quick Reference

import scitex as stx

# Create figure (returns FigWrapper, AxisWrapper)
fig, ax = stx.plt.subplots()
fig, axes = stx.plt.subplots(2, 2)

# Data-tracking plot methods (all record data for CSV export)
ax.plot_line(x, y, label="signal")
ax.plot_scatter(x, y, alpha=0.5)
ax.plot_bar(labels, values)
ax.plot_heatmap(matrix)
ax.plot_violin(groups)
ax.plot_box(groups)
ax.plot_kde(data)

# Axis helper (xlabel, ylabel, title in one call)
ax.set_xyt("Time (s)", "Amplitude (mV)", "EEG Signal")

# Save (auto-exports CSV + YAML recipe)
stx.io.save(fig, "figure.png")

Plot Types (47)

All ax.plot_* methods track their input data so it can be exported as CSV.

Category	Types
Line & Curve	`plot`, `step`, `fill`, `fill_between`, `fill_betweenx`, `errorbar`, `stackplot`, `stairs`
Scatter & Points	`scatter`
Bar & Categorical	`bar`, `barh`
Distribution	`hist`, `hist2d`, `boxplot`, `violinplot`, `ecdf`
2D Image & Matrix	`imshow`, `matshow`, `pcolor`, `pcolormesh`, `hexbin`, `spy`
Contour & Surface	`contour`, `contourf`, `tricontour`, `tricontourf`, `tripcolor`, `triplot`
Spectral & Signal	`specgram`, `psd`, `csd`, `cohere`, `angle_spectrum`, `magnitude_spectrum`, `phase_spectrum`, `acorr`, `xcorr`
Vector & Flow	`quiver`, `barbs`, `streamplot`
Special	`pie`, `stem`, `eventplot`, `loglog`, `semilogx`, `semilogy`, `graph`

Standard matplotlib methods (ax.plot, ax.scatter, etc.) also work and are tracked.

MM-Based Layout

Figures use millimeter dimensions for precise control:

fig, ax = stx.plt.subplots(
    axes_width_mm=80,
    axes_height_mm=60,
)

# Multi-panel with exact spacing
fig, axes = stx.plt.subplots(
    nrows=2, ncols=3,
    axes_width_mm=50,
    axes_height_mm=40,
)

Axis Helpers

# Set xlabel, ylabel, title in one call
ax.set_xyt("X Label", "Y Label", "Title")

# Automatic axis label formatting with units
ax.set_xyt("Frequency (Hz)", "Power (dB)", "Power Spectrum")

Figure Composition

Combine multiple figures into panels:

# Compose panels A, B, C
stx.plt.compose(
    sources=["fig_a.png", "fig_b.png", "fig_c.png"],
    output_path="combined.png",
    layout="horizontal",
    panel_labels=True,
)

Statistical Annotations

Add significance brackets to figures:

# In the declarative spec
stat_annotations:
  - x1: 0
    x2: 1
    y: 5.5
    text: "***"
  - x1: 0
    x2: 2
    y: 6.5
    p_value: 0.02
    style: stars

Reproducible Figures

When saved via stx.io.save, figures get a .yaml recipe. Reproduce any figure from its recipe:

stx.plt.reproduce("figure.yaml", output_path="reproduced.png")

Style Presets

import figrecipe as fr

fr.load_style("SCITEX")       # Publication defaults
fr.load_style("MATPLOTLIB")   # Standard matplotlib

See the Plot Gallery for visual examples.

API Reference

See scitex.plt API Reference for the auto-generated Python API.

PLT Module (stx.plt)