ambivalent 0.1.0

Minimal, beautiful (+ highly-customizable) styles for Matplotlib.

ambivalent 🤷🏻‍♂️

Sam Foreman 2023-12-29

opinionated $\longrightarrow$ ambivalent

Clean, simple style for Matplotlib figures.

Transparent backgrounds with grey text $\textcolor{#838383}{\blacksquare}$ that are accessible / legible and {light, dark}-mode independent.

Examples[^1]

More Examples

Even more…

Figure 9: Figure from l2hmc-qcd Notebook

InferenceData

Figure 10: $\varepsilon_{x}$ during training

Figure 11: $\varepsilon_{x}$ during training

Figure 12: Figure from l2hmc-qcd Notebook

Getting Started

• Use ambivalend.STYLES['ambivalent'] as the default style for matplotlib.

import matplotlib.pyplot as plt
from ambivalent import STYLES
plt.style.use(STYLES['ambivalent'])

seaborn Tips Dataset

• Tips Dataset Example
• Seaborn Gallery

View Code

import seaborn as sns

tips = sns.load_dataset("tips")
tips.head()

fig, ax = plt.subplots(figsize=(6, 6))  # , ncols=2)

_ = sns.kdeplot(
   data=tips, x="total_bill", hue="size",
   fill=True, common_norm=False, palette="flare_r",
   alpha=.3, linewidth=0,
   ax=ax,  # [0],
)
_ = ax.set_ylabel('')
# _ = sns.kdeplot(
#    data=tips, x="tip", hue="size",
#    fill=True, common_norm=False, palette="crest",
#    alpha=.3, linewidth=0,
#    ax=ax[1],
# )

seaborn Scatter Plot

View Code

import seaborn as sns
import matplotlib.pyplot as plt

# Load the example diamonds dataset
diamonds = sns.load_dataset("diamonds")

# Draw a scatter plot while assigning point colors and sizes to different
# variables in the dataset
f, ax = plt.subplots(figsize=(6, 6))
_ = sns.despine(f, left=True, bottom=True)
_ = clarity_ranking = ["I1", "SI2", "SI1", "VS2", "VS1", "VVS2", "VVS1", "IF"]
_ = sns.scatterplot(x="carat", y="price",
                hue="clarity", size="depth",
                palette="flare",
                hue_order=clarity_ranking,
                sizes=(1, 8), linewidth=0,
                data=diamonds, ax=ax)

Histogram + Scatter Plot

View Code

import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt

# Simulate data from a bivariate Gaussian
n = 10000
mean = [0, 0]
cov = [(2, .4), (.4, .2)]
rng = np.random.RandomState(0)
x, y = rng.multivariate_normal(mean, cov, n).T

# Draw a combo histogram and scatterplot with density contours
f, ax = plt.subplots(figsize=(6, 6))
_ = sns.scatterplot(x=x, y=y, s=5, color="#666666", alpha=0.3)
_ = sns.histplot(x=x, y=y, bins=50, pthresh=.1, cmap="flare_r")
_ = sns.kdeplot(x=x, y=y, levels=5, color="w", linewidths=1)
_ = ax.set_xlabel('x')
_ = ax.set_ylabel('y')
_ = plt.show()

Jointplot

View Code

import seaborn as sns
# Load the penguins dataset
penguins = sns.load_dataset("penguins")
# Show the joint distribution using kernel density estimation
import matplotlib as mpl
with mpl.rc_context(plt.rcParams.update({'axes.grid': False})):
  g = sns.jointplot(
      data=penguins,
      x="bill_length_mm",
      y="bill_depth_mm",
      hue="species",
      # kind="kde",
      edgecolors='none',
      alpha=0.4,
  )
  _ = plt.grid(False)
  plt.show()

Matplotlib Histograms

View Code

import matplotlib.pyplot as plt
import numpy as np

n_bins = 10
x = np.random.randn(1000, 3)

plt.rcParams['axes.grid'] = True

fig, ((ax0, ax1), (ax2, ax3)) = plt.subplots(nrows=2, ncols=2)

colors = ['#333333', '#666666', '#999999']
ax0.hist(x, n_bins, density=True, histtype='bar', color=colors, label=colors)
_ = ax0.legend()
_ = ax0.set_title('bars with legend')

_ = ax1.hist(x, n_bins, density=True, histtype='bar', stacked=True, alpha=0.4)
_ = ax1.set_title('stacked bar')

_ = ax2.hist(x, n_bins, histtype='step', stacked=True, fill=False)
_ = ax2.set_title('stack step (unfilled)')

# Make a multiple-histogram of data-sets with different length.
x_multi = [np.random.randn(n) for n in [10000, 5000, 2000]]
_ = ax3.hist(x_multi, n_bins, histtype='bar')
_ = ax3.set_title('different sample sizes')

_ = fig.tight_layout()
plt.show()

Status

Last Updated: 12/29/2023 @ 10:42:06

[^1]: Examples from Matplotlib Examples