Statistical Visualization

MinexPy includes a minexpy.statviz module for common statistical plots used in geoscience QA/QC and exploratory analysis.

Available Plot Functions

• plot_histogram: Histogram with linear or log x-axis.
• plot_box_violin: Box or violin representation for one or multiple variables.
• plot_ecdf: Empirical cumulative distribution function.
• plot_qq: Quantile-quantile diagnostic against a theoretical distribution.
• plot_pp: Probability-probability diagnostic against a theoretical distribution.
• plot_scatter: Scatter plot with optional linear trend line.

All plot functions set axis labels and return (figure, axis) for further customization.

plot_histogram

Create histograms in linear and log scale using the same Zn dataset:

import matplotlib.pyplot as plt
import numpy as np
from minexpy.statviz import plot_histogram

rng = np.random.default_rng(42)
zn = rng.lognormal(mean=2.2, sigma=0.35, size=250)

fig, ax = plot_histogram(zn, bins=30, scale="linear", xlabel="Zn (ppm)")
plt.show()

fig, ax = plot_histogram(zn, bins=30, scale="log", xlabel="Zn (ppm, log scale)")
plt.show()

plot_box_violin

Visualize Zn and Cu distributions as box and violin plots:

import matplotlib.pyplot as plt
import numpy as np
from minexpy.statviz import plot_box_violin

rng = np.random.default_rng(42)
zn = rng.lognormal(mean=2.2, sigma=0.35, size=250)
cu = 0.35 * zn + rng.normal(0, 1.5, size=250)

fig, ax = plot_box_violin({"Zn": zn, "Cu": cu}, kind="box", ylabel="Concentration (ppm)")
plt.show()

fig, ax = plot_box_violin({"Zn": zn, "Cu": cu}, kind="violin", ylabel="Concentration (ppm)")
plt.show()

plot_ecdf

Compare empirical cumulative distributions for Zn and Cu:

import matplotlib.pyplot as plt
import numpy as np
from minexpy.statviz import plot_ecdf

rng = np.random.default_rng(42)
zn = rng.lognormal(mean=2.2, sigma=0.35, size=250)
cu = 0.35 * zn + rng.normal(0, 1.5, size=250)

fig, ax = plot_ecdf({"Zn": zn, "Cu": cu}, xlabel="Concentration (ppm)")
plt.show()

plot_qq

Check Zn sample quantiles against a normal reference:

import matplotlib.pyplot as plt
import numpy as np
from minexpy.statviz import plot_qq

rng = np.random.default_rng(42)
zn = rng.lognormal(mean=2.2, sigma=0.35, size=250)

fig, ax = plot_qq(zn, distribution="norm", ylabel="Zn Sample Quantiles")
plt.show()

plot_pp

Compare empirical vs theoretical cumulative probabilities:

import matplotlib.pyplot as plt
import numpy as np
from minexpy.statviz import plot_pp

rng = np.random.default_rng(42)
zn = rng.lognormal(mean=2.2, sigma=0.35, size=250)

fig, ax = plot_pp(zn, distribution="norm", ylabel="Empirical Probability")
plt.show()

plot_scatter

Plot bivariate Zn-Cu relation with a fitted trend line:

import matplotlib.pyplot as plt
import numpy as np
from minexpy.statviz import plot_scatter

rng = np.random.default_rng(42)
zn = rng.lognormal(mean=2.2, sigma=0.35, size=250)
cu = 0.35 * zn + rng.normal(0, 1.5, size=250)

fig, ax = plot_scatter(
    zn,
    cu,
    add_trendline=True,
    xlabel="Zn (ppm)",
    ylabel="Cu (ppm)",
    title="Zn vs Cu",
)
plt.show()