plotlyhep

The mplhep look for Plotly: CMS and ATLAS styles, experiment labels, histogram and ratio-panel helpers, per-bin hover, one-line embedding into HTML slides, and a converter between matplotlib and Plotly figures. Fidelity is checked by pixel diff against mplhep and against ROOT's own tutorial output, not asserted.

import numpy as np
import plotlyhep as php

php.style.use("CMS")
bins = np.linspace(0, 200, 41)
mc = np.histogram(np.random.default_rng(1).normal(90, 25, 3000), bins)[0] * 1.3
data = np.histogram(np.random.default_rng(2).normal(90, 25, 4000), bins)[0]

fig = php.figure()
php.histplot(fig, mc, bins, label="MC")
php.histplot(fig, data, bins, yerr=True, histtype="errorbar", color="black", label="Data")
php.cms.label(fig, "Preliminary", data=True, lumi=138, com=13.6)
php.set_xlabel(fig, "m<sub>jj</sub> [GeV]")
php.set_ylabel(fig, "Events")
fig.show()

That figure, live (hover a bin):

editsave

Where to go

• Install — from git, a tag or a release tarball.
• Guide — every feature with an example.
• Gallery — ATLAS and CMS Open Data figures with hover that explains each process, rebuilt to match ROOT's tutorial output, and an editor that loads any of them or your own Plotly JSON.
• API reference — every public function.

What it mirrors

mplhep	plotlyhep
hep.style.use("CMS")	php.style.use("CMS")
plt.subplots(figsize=hep.style.CMS["figure.figsize"])	php.figure("CMS") (1000×1000 px, the same size at 100 dpi)
hep.histplot(H, bins, histtype=..., yerr=..., stack=...)	php.histplot(fig, H, bins, ...)
hep.hist2dplot(H, xbins, ybins)	php.hist2dplot(fig, H, xbins, ybins)
hep.cms.label("Preliminary", data=True, lumi=138, com=13.6, loc=0)	php.cms.label(fig, "Preliminary", data=True, lumi=138, com=13.6, loc=0)
ax.set_xlabel(..., loc="right")	php.set_xlabel(fig, ...)
plt.subplots(2, 1, gridspec_kw=dict(height_ratios=(3, 1), hspace=0.05))	php.ratio_figure("CMS", height_ratios=(3, 1), hspace=0.05)