wehub-resource-sync
8cb1f9f479
Publish SDK (PyPI) / publish (push) Has been cancelled
Publish SDK (npm) / publish (@aionui/officecli-sdk) (push) Has been cancelled
SDK smoke / smoke (windows-latest) (push) Has been cancelled
Publish SDK (npm) / publish (@officecli/officecli-sdk) (push) Has been cancelled
Publish SDK (npm) / publish (@officecli/sdk) (push) Has been cancelled
Publish SDK (npm) / publish (officecli-sdk) (push) Has been cancelled
SDK smoke / smoke (macos-latest) (push) Has been cancelled
SDK smoke / smoke (ubuntu-latest) (push) Has been cancelled
Skill parity / diff (push) Has been cancelled
561 lines
29 KiB
Python
561 lines
29 KiB
Python
#!/usr/bin/env python3
|
||
"""
|
||
Histogram Charts — Grand Showcase
|
||
==================================
|
||
|
||
The most thorough, most visually polished histogram demo officecli can
|
||
produce. Every binning knob, every styling vocabulary, every canonical
|
||
distribution shape, six design themes on one dataset, four font type
|
||
specimens, and a cohesive production-grade ML dashboard — all driven by
|
||
the real officecli Python SDK.
|
||
|
||
SDK twin of charts-histogram.sh (officecli CLI). Both produce an equivalent
|
||
charts-histogram.xlsx. This one drives the **officecli Python SDK**
|
||
(`pip install officecli-sdk`): one resident is started and every sheet and
|
||
chart is shipped over the named pipe in `doc.batch(...)` round-trips. Each
|
||
item is the same `{"command","parent","type","props"}` dict you'd put in an
|
||
`officecli batch` list.
|
||
|
||
Generates: charts-histogram.xlsx (6 sheets, 29 histograms)
|
||
|
||
0-Hero 1 magazine-grade full-bleed hero poster chart
|
||
1-Binning Lab 6 charts — every binning knob, identical styling
|
||
2-Distribution Zoo 6 canonical real-world distribution shapes
|
||
3-Theme Gallery 6 design themes on the SAME dataset
|
||
4-Typography 4 font-family type specimens
|
||
5-ML Dashboard 6-chart "Production ML Model Report" dashboard
|
||
|
||
Usage:
|
||
pip install officecli-sdk # plus the `officecli` binary on PATH
|
||
python3 charts-histogram.py
|
||
"""
|
||
|
||
import os
|
||
import sys
|
||
import random
|
||
import math
|
||
|
||
# --- locate the SDK: prefer an installed `officecli-sdk`, else the in-repo copy
|
||
try:
|
||
import officecli # pip install officecli-sdk
|
||
except ImportError:
|
||
sys.path.insert(0, os.path.join(os.path.dirname(os.path.abspath(__file__)),
|
||
"..", "..", "..", "sdk", "python"))
|
||
import officecli
|
||
|
||
FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "charts-histogram.xlsx")
|
||
|
||
|
||
# --------------------------------------------------------------------------
|
||
# Batch-item helpers — each returns one dict you'd put in an `officecli batch`
|
||
# list. The SDK ships the whole list in a single named-pipe round-trip.
|
||
# --------------------------------------------------------------------------
|
||
def add_sheet(name):
|
||
"""One `add sheet` item in batch-shape."""
|
||
return {"command": "add", "parent": "/", "type": "sheet", "props": {"name": name}}
|
||
|
||
|
||
def chart(parent, **props):
|
||
"""One `add chart` item in batch-shape."""
|
||
props.setdefault("chartType", "histogram")
|
||
return {"command": "add", "parent": parent, "type": "chart", "props": props}
|
||
|
||
|
||
# --------------------------------------------------------------------------
|
||
# Deterministic sample generators — same seed, same file every regeneration.
|
||
# All datasets are CSV-joined once here and reused across sheets.
|
||
# --------------------------------------------------------------------------
|
||
def csv(values):
|
||
return ",".join(str(v) for v in values)
|
||
|
||
# The "reference" bell curve — 200 samples around 75±12. Used by the hero,
|
||
# the binning lab, the theme gallery, the typography specimens, and the zoo.
|
||
random.seed(42)
|
||
BELL_200 = sorted(round(random.gauss(75, 12), 1) for _ in range(200))
|
||
BELL_CSV = csv(BELL_200)
|
||
|
||
# Bimodal: two cohorts (beginners ~55, experts ~88) glued together.
|
||
random.seed(7)
|
||
BIMODAL = sorted(
|
||
[round(random.gauss(55, 6), 1) for _ in range(80)]
|
||
+ [round(random.gauss(88, 5), 1) for _ in range(80)]
|
||
)
|
||
BIMODAL_CSV = csv(BIMODAL)
|
||
|
||
# Right-skewed / log-normal: classic income shape.
|
||
random.seed(11)
|
||
LOGNORM = sorted(round(math.exp(random.gauss(3.2, 0.55)), 1) for _ in range(180))
|
||
LOGNORM_CSV = csv(LOGNORM)
|
||
|
||
# Left-skewed: retirement ages — most cluster high, a few retire early.
|
||
random.seed(23)
|
||
LEFT_SKEW = sorted(round(75 - math.exp(random.gauss(1.6, 0.6)), 1) for _ in range(140))
|
||
LEFT_CSV = csv(LEFT_SKEW)
|
||
|
||
# Uniform: random draws evenly distributed across a range.
|
||
random.seed(31)
|
||
UNIFORM = sorted(round(random.uniform(0, 100), 1) for _ in range(160))
|
||
UNIFORM_CSV = csv(UNIFORM)
|
||
|
||
# Heavy-tailed (Pareto): most small, tiny fraction catastrophic.
|
||
random.seed(47)
|
||
PARETO = sorted(round(random.paretovariate(1.6) * 20, 1) for _ in range(200))
|
||
PARETO_CSV = csv(PARETO)
|
||
|
||
# --- ML Dashboard datasets (sheet 5) ---
|
||
random.seed(101)
|
||
LATENCY_MS = sorted(round(random.paretovariate(1.8) * 15 + 10, 1) for _ in range(250))
|
||
LATENCY_CSV = csv(LATENCY_MS)
|
||
|
||
random.seed(102)
|
||
CONFIDENCE = sorted(round(random.betavariate(6, 2) * 100, 2) for _ in range(240))
|
||
CONFIDENCE_CSV = csv(CONFIDENCE)
|
||
|
||
random.seed(103)
|
||
ERROR_MAG = sorted(round(abs(random.gauss(0, 1.5)), 3) for _ in range(180))
|
||
ERROR_MAG_CSV = csv(ERROR_MAG)
|
||
|
||
random.seed(104)
|
||
TOKEN_LEN = sorted(
|
||
[max(1, round(random.gauss(180, 40))) for _ in range(100)]
|
||
+ [max(1, round(random.gauss(520, 90))) for _ in range(80)]
|
||
)
|
||
TOKEN_CSV = csv(TOKEN_LEN)
|
||
|
||
random.seed(105)
|
||
GPU_UTIL = sorted(round(min(99.0, max(30.0, random.gauss(82, 8))), 1) for _ in range(200))
|
||
GPU_CSV = csv(GPU_UTIL)
|
||
|
||
random.seed(106)
|
||
COST_REQ = sorted(round(math.exp(random.gauss(-3.2, 0.9)) * 1000, 3) for _ in range(220))
|
||
COST_CSV = csv(COST_REQ)
|
||
|
||
|
||
print(f"Building {FILE} ...")
|
||
|
||
with officecli.create(FILE, "--force") as doc:
|
||
|
||
# ======================================================================
|
||
# Sheet 0: "0-Hero" — the full-bleed magazine hero poster
|
||
#
|
||
# A single giant chart using EVERY histogram knob at once:
|
||
# - Dark "Midnight Academia" palette: navy plot area, gold bars, cream text
|
||
# - title.* (color/size/bold/font/shadow)
|
||
# - series.shadow + fill
|
||
# - axisline + axisfont + axisTitle.*
|
||
# - plotareafill / plotarea.border / chartareafill / chartarea.border
|
||
# - axismin / axismax / majorunit (locked Y scale)
|
||
# - gridlineColor
|
||
# - dataLabels + datalabels.numfmt
|
||
# - legend=top + legend.overlay + legendfont
|
||
# - intervalClosed=l + explicit binCount
|
||
#
|
||
# This chart is the "representative sample" — if it renders correctly, the
|
||
# entire histogram pipeline is healthy.
|
||
# ======================================================================
|
||
print("\n--- 0-Hero ---")
|
||
doc.batch([
|
||
# rename the default Sheet1 → "0-Hero"
|
||
{"command": "set", "path": "/Sheet1", "props": {"name": "0-Hero"}},
|
||
# EVERY knob — title/series/axis/plotarea/chartarea/shadow/scaling/legend/datalabel
|
||
chart("/0-Hero",
|
||
title="The Shape of Data · 200-sample bell curve",
|
||
**{"title.color": "F5F1E0", "title.size": "22", "title.bold": "true",
|
||
"title.font": "Helvetica Neue",
|
||
"title.shadow": "000000-8-45-4-70"},
|
||
series1=f"Samples:{BELL_CSV}",
|
||
binCount="24", intervalClosed="l",
|
||
fill="F0C96A", **{"series.shadow": "000000-8-45-4-60"},
|
||
axismin="0", axismax="28", majorunit="4",
|
||
xAxisTitle="Score", yAxisTitle="Frequency",
|
||
**{"axisTitle.color": "C9B87A", "axisTitle.size": "13",
|
||
"axisTitle.bold": "true", "axisTitle.font": "Helvetica Neue",
|
||
"axisfont": "10:B8B090:Helvetica Neue",
|
||
"axisline": "6A6448:1.5"},
|
||
gridlineColor="2F3544",
|
||
plotareafill="1A1F2C", **{"plotarea.border": "3A3E4E:1.25"},
|
||
chartareafill="0B0F18", **{"chartarea.border": "2A2E3E:1"},
|
||
dataLabels="true", **{"datalabels.numfmt": "0"},
|
||
legend="top", **{"legend.overlay": "false",
|
||
"legendfont": "11:D4C994:Helvetica Neue"},
|
||
x="0", y="0", width="27", height="38"),
|
||
])
|
||
|
||
# ======================================================================
|
||
# Sheet 1: "1-Binning Lab"
|
||
#
|
||
# Six histograms, SAME dataset (BELL_200), IDENTICAL typography / colors /
|
||
# frames — the ONLY thing that varies is the binning strategy. Put side by
|
||
# side, this sheet is the "Rosetta stone": once you see how each binning
|
||
# knob reshapes the bars, you'll never be confused about which to use.
|
||
#
|
||
# ┌──────────┬──────────┐
|
||
# │ 1. auto │ 2. count │
|
||
# ├──────────┼──────────┤
|
||
# │ 3. fine │ 4. width │
|
||
# ├──────────┼──────────┤
|
||
# │ 5. fence │ 6. lclos │
|
||
# └──────────┴──────────┘
|
||
# ======================================================================
|
||
print("\n--- 1-Binning Lab ---")
|
||
|
||
# Shared "clean lab" style — every chart on this sheet wears the exact same
|
||
# outfit so the bin-shape difference is the only visible variable.
|
||
LAB = {
|
||
"fill": "4472C4",
|
||
"title.color": "1F2937", "title.size": "13", "title.bold": "true",
|
||
"title.font": "Helvetica Neue",
|
||
"xAxisTitle": "Score", "yAxisTitle": "Count",
|
||
"axisTitle.color": "6B7280", "axisTitle.size": "10",
|
||
"axisTitle.font": "Helvetica Neue",
|
||
"axisfont": "9:6B7280:Helvetica Neue",
|
||
"gridlineColor": "F0F0F0",
|
||
"plotareafill": "FFFFFF", "plotarea.border": "E5E7EB:0.75",
|
||
"chartareafill": "F9FAFB", "chartarea.border": "E5E7EB:0.75",
|
||
"axisline": "9CA3AF:0.75",
|
||
}
|
||
|
||
doc.batch([
|
||
add_sheet("1-Binning Lab"),
|
||
# no binCount, no binSize — Excel picks the bin count automatically.
|
||
chart("/1-Binning Lab", title="1 · Auto-binning (Excel default)",
|
||
series1=f"Samples:{BELL_CSV}", **LAB,
|
||
x="0", y="0", width="13", height="18"),
|
||
# binCount=8 — coarse. Fewer, wider bars. Good for "what's the mode?"
|
||
chart("/1-Binning Lab", title="2 · binCount=8 (coarse)",
|
||
series1=f"Samples:{BELL_CSV}", binCount="8", **LAB,
|
||
x="14", y="0", width="13", height="18"),
|
||
# binCount=32 — fine. Many narrow bars. Good for "is it really Gaussian?"
|
||
chart("/1-Binning Lab", title="3 · binCount=32 (fine)",
|
||
series1=f"Samples:{BELL_CSV}", binCount="32", **LAB,
|
||
x="0", y="19", width="13", height="18"),
|
||
# binSize=5 — fixed bin width. Use when you want human-friendly bin
|
||
# boundaries (multiples of 5, 10, etc) regardless of data range.
|
||
chart("/1-Binning Lab", title="4 · binSize=5 (fixed-width bins)",
|
||
series1=f"Samples:{BELL_CSV}", binSize="5", **LAB,
|
||
x="14", y="19", width="13", height="18"),
|
||
# underflowBin=55 + overflowBin=95 — outlier fencing. Everything below
|
||
# 55 or above 95 collapses into a single <55 / >95 bar.
|
||
chart("/1-Binning Lab", title="5 · underflow=55 · overflow=95 (fencing)",
|
||
series1=f"Samples:{BELL_CSV}", binSize="5", underflowBin="55",
|
||
overflowBin="95", **LAB,
|
||
x="0", y="38", width="13", height="18"),
|
||
# intervalClosed=l (half-open [a,b)) + gapWidth=30 — shows the
|
||
# "left-closed" variant AND pushes bars apart so you can see each one.
|
||
# Useful when the dataset has values lying exactly on a bin boundary.
|
||
chart("/1-Binning Lab", title="6 · [a,b) intervals + gapWidth=30",
|
||
series1=f"Samples:{BELL_CSV}", binCount="16", intervalClosed="l",
|
||
gapWidth="30", **LAB,
|
||
x="14", y="38", width="13", height="18"),
|
||
])
|
||
|
||
# ======================================================================
|
||
# Sheet 2: "2-Distribution Zoo"
|
||
#
|
||
# A cohesive 2x3 gallery of the canonical distribution shapes you'll see
|
||
# in production data. Pattern recognition: if you ever see one of these
|
||
# shapes in a telemetry chart, you know immediately what's going on.
|
||
#
|
||
# Every chart shares the same typography + plot/chart area frames; only
|
||
# the fill color and data change. Uses different binning strategies
|
||
# appropriate to each distribution.
|
||
# ======================================================================
|
||
print("\n--- 2-Distribution Zoo ---")
|
||
|
||
ZOO = {
|
||
"title.color": "1F2937", "title.size": "13", "title.bold": "true",
|
||
"title.font": "Helvetica Neue",
|
||
"axisTitle.color": "6B7280", "axisTitle.size": "10",
|
||
"axisTitle.font": "Helvetica Neue",
|
||
"axisfont": "9:6B7280:Helvetica Neue",
|
||
"gridlineColor": "EFEFEF",
|
||
"plotareafill": "FFFFFF", "plotarea.border": "E5E7EB:0.75",
|
||
"chartareafill": "F9FAFB", "chartarea.border": "E5E7EB:0.75",
|
||
"axisline": "9CA3AF:0.75",
|
||
}
|
||
|
||
doc.batch([
|
||
add_sheet("2-Distribution Zoo"),
|
||
# classic bell curve reference, binCount=18, midnight blue fill.
|
||
chart("/2-Distribution Zoo", title="Normal · bell curve (reference)",
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="2F5597",
|
||
xAxisTitle="Score", yAxisTitle="Count", **ZOO,
|
||
x="0", y="0", width="13", height="18"),
|
||
# bimodal — two hidden populations. Narrow bins reveal the split.
|
||
chart("/2-Distribution Zoo", title="Bimodal · two hidden cohorts",
|
||
series1=f"Score:{BIMODAL_CSV}", binCount="22", fill="ED7D31",
|
||
xAxisTitle="Test score", yAxisTitle="Students", **ZOO,
|
||
x="14", y="0", width="13", height="18"),
|
||
# right-skewed log-normal. Mean >> median, long tail to the right.
|
||
chart("/2-Distribution Zoo", title="Right-skewed · log-normal (income)",
|
||
series1=f"Income:{LOGNORM_CSV}", binCount="20", fill="70AD47",
|
||
xAxisTitle="Monthly income ($k)", yAxisTitle="People", **ZOO,
|
||
x="0", y="19", width="13", height="18"),
|
||
# left-skewed — retirement ages cluster high, tail stretches left.
|
||
chart("/2-Distribution Zoo", title="Left-skewed · retirement ages",
|
||
series1=f"Age:{LEFT_CSV}", binCount="18", fill="7030A0",
|
||
xAxisTitle="Age at retirement", yAxisTitle="Retirees", **ZOO,
|
||
x="14", y="19", width="13", height="18"),
|
||
# uniform — every value equally likely. binSize emphasizes the
|
||
# "flat floor" visual tell.
|
||
chart("/2-Distribution Zoo", title="Uniform · flat floor",
|
||
series1=f"Draws:{UNIFORM_CSV}", binSize="10", fill="00B0F0",
|
||
xAxisTitle="Random draw (0-100)", yAxisTitle="Count", **ZOO,
|
||
x="0", y="38", width="13", height="18"),
|
||
# heavy-tailed Pareto + overflowBin. Fences the catastrophic tail so
|
||
# the interesting bulk of the distribution stays readable.
|
||
chart("/2-Distribution Zoo", title="Heavy-tailed · Pareto (overflow=250)",
|
||
series1=f"Latency:{PARETO_CSV}", binSize="20", overflowBin="250",
|
||
fill="C00000",
|
||
xAxisTitle="Latency (ms)", yAxisTitle="Requests", **ZOO,
|
||
x="14", y="38", width="13", height="18"),
|
||
])
|
||
|
||
# ======================================================================
|
||
# Sheet 3: "3-Theme Gallery"
|
||
#
|
||
# Six complete design themes applied to the SAME bell-curve dataset. Each
|
||
# theme is a coordinated palette: plot-area fill, chart-area fill, series
|
||
# fill, gridline color, axis line color, tick-label color, title color,
|
||
# title font — all chosen to read as one coherent mood.
|
||
#
|
||
# Grid:
|
||
# ┌─────────────┬─────────────┐
|
||
# │ 1. Midnight │ 2. Sunset │
|
||
# ├─────────────┼─────────────┤
|
||
# │ 3. Forest │ 4. Mono │
|
||
# ├─────────────┼─────────────┤
|
||
# │ 5. Neon │ 6. Pastel │
|
||
# └─────────────┴─────────────┘
|
||
# ======================================================================
|
||
print("\n--- 3-Theme Gallery ---")
|
||
doc.batch([
|
||
add_sheet("3-Theme Gallery"),
|
||
# Theme 1 · Midnight Academia — dark plot area, gold bars, shadows
|
||
chart("/3-Theme Gallery", title="Midnight Academia",
|
||
**{"title.color": "F5F1E0", "title.size": "14", "title.bold": "true",
|
||
"title.font": "Georgia", "title.shadow": "000000-6-45-3-70"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="F0C96A",
|
||
**{"series.shadow": "000000-6-45-3-55"},
|
||
plotareafill="1A1F2C", **{"plotarea.border": "3A3E4E:1"},
|
||
chartareafill="0B0F18", **{"chartarea.border": "2A2E3E:0.75"},
|
||
gridlineColor="2F3544",
|
||
**{"axisfont": "9:B8B090:Georgia"},
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "C9B87A", "axisTitle.size": "10",
|
||
"axisTitle.font": "Georgia", "axisline": "5A5848:1"},
|
||
x="0", y="0", width="13", height="18"),
|
||
# Theme 2 · Sunset Terracotta (warm cream + coral, serif)
|
||
chart("/3-Theme Gallery", title="Sunset Terracotta",
|
||
**{"title.color": "3F2818", "title.size": "14", "title.bold": "true",
|
||
"title.font": "Georgia"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="E85D4A",
|
||
plotareafill="FFF5E8", **{"plotarea.border": "F0D8B0:1"},
|
||
chartareafill="FFE6C7", **{"chartarea.border": "E6BC88:1"},
|
||
gridlineColor="F5C98A",
|
||
**{"axisfont": "9:6B4A2A:Georgia"},
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "A8522C", "axisTitle.size": "10",
|
||
"axisTitle.font": "Georgia", "axisline": "C08050:1"},
|
||
x="14", y="0", width="13", height="18"),
|
||
# Theme 3 · Forest Parchment (beige + forest green, serif)
|
||
chart("/3-Theme Gallery", title="Forest Parchment",
|
||
**{"title.color": "1F3A1F", "title.size": "14", "title.bold": "true",
|
||
"title.font": "Georgia"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="2F5D3A",
|
||
plotareafill="F3EDD8", **{"plotarea.border": "C8B890:1"},
|
||
chartareafill="EADFBE", **{"chartarea.border": "A89858:1"},
|
||
gridlineColor="C0B888",
|
||
**{"axisfont": "9:4A5A3A:Georgia"},
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "3F5A2F", "axisTitle.size": "10",
|
||
"axisTitle.font": "Georgia", "axisline": "6A7A4A:1"},
|
||
x="0", y="19", width="13", height="18"),
|
||
# Theme 4 · Editorial Mono (pure grayscale, sans)
|
||
chart("/3-Theme Gallery", title="Editorial Mono",
|
||
**{"title.color": "111111", "title.size": "14", "title.bold": "true",
|
||
"title.font": "Helvetica Neue"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="2A2A2A",
|
||
plotareafill="FFFFFF", **{"plotarea.border": "CCCCCC:0.75"},
|
||
chartareafill="FAFAFA", **{"chartarea.border": "E0E0E0:0.75"},
|
||
gridlineColor="EEEEEE",
|
||
**{"axisfont": "9:555555:Helvetica Neue"},
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "333333", "axisTitle.size": "10",
|
||
"axisTitle.font": "Helvetica Neue", "axisline": "888888:1"},
|
||
x="14", y="19", width="13", height="18"),
|
||
# Theme 5 · Neon Terminal (black + electric cyan, mono)
|
||
chart("/3-Theme Gallery", title="Neon Terminal",
|
||
**{"title.color": "00F0C8", "title.size": "14", "title.bold": "true",
|
||
"title.font": "Courier New", "title.shadow": "00F0C8-6-45-0-40"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="00F0C8",
|
||
**{"series.shadow": "00F0C8-8-45-0-45"},
|
||
plotareafill="0A0A14", **{"plotarea.border": "1F2F3F:1"},
|
||
chartareafill="000008", **{"chartarea.border": "1F1F2F:1"},
|
||
gridlineColor="1A2A3A",
|
||
**{"axisfont": "9:00D0E8:Courier New"},
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "00D0E8", "axisTitle.size": "10",
|
||
"axisTitle.font": "Courier New", "axisline": "00707F:1"},
|
||
x="0", y="38", width="13", height="18"),
|
||
# Theme 6 · Pastel Bloom (lavender cream + rose, sans)
|
||
chart("/3-Theme Gallery", title="Pastel Bloom",
|
||
**{"title.color": "5A3C4A", "title.size": "14", "title.bold": "true",
|
||
"title.font": "Helvetica Neue"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="F5A7C8",
|
||
plotareafill="FDF4F8", **{"plotarea.border": "F0D0E0:1"},
|
||
chartareafill="FAEDF2", **{"chartarea.border": "F0C0D8:1"},
|
||
gridlineColor="F5D8E5",
|
||
**{"axisfont": "9:8A6878:Helvetica Neue"},
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "A04C6A", "axisTitle.size": "10",
|
||
"axisTitle.font": "Helvetica Neue", "axisline": "C888A0:1"},
|
||
x="14", y="38", width="13", height="18"),
|
||
])
|
||
|
||
# ======================================================================
|
||
# Sheet 4: "4-Typography"
|
||
#
|
||
# Four font-family "type specimens". Same data, same geometry, same colors —
|
||
# only the font varies. Side-by-side, this shows how typography alone reads
|
||
# as tone: Helvetica is corporate, Georgia is editorial, Courier is data,
|
||
# Verdana is approachable.
|
||
# ======================================================================
|
||
print("\n--- 4-Typography ---")
|
||
doc.batch([
|
||
add_sheet("4-Typography"),
|
||
# Specimen 1 · Helvetica Neue (modern sans — dashboards, corporate reports)
|
||
chart("/4-Typography", title="Helvetica Neue · modern sans",
|
||
**{"title.color": "1F2937", "title.size": "16", "title.bold": "true",
|
||
"title.font": "Helvetica Neue"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="4472C4",
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "4472C4", "axisTitle.size": "11",
|
||
"axisTitle.font": "Helvetica Neue",
|
||
"axisfont": "10:6B7280:Helvetica Neue"},
|
||
gridlineColor="EEEEEE",
|
||
plotareafill="FFFFFF", **{"plotarea.border": "E5E7EB:0.75"},
|
||
chartareafill="F9FAFB", **{"chartarea.border": "E5E7EB:0.75"},
|
||
x="0", y="0", width="13", height="18"),
|
||
# Specimen 2 · Georgia (editorial serif — magazines, long-form reports)
|
||
chart("/4-Typography", title="Georgia · editorial serif",
|
||
**{"title.color": "3F2818", "title.size": "16", "title.bold": "true",
|
||
"title.font": "Georgia"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="A8522C",
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "A8522C", "axisTitle.size": "11",
|
||
"axisTitle.font": "Georgia",
|
||
"axisfont": "10:6B4A2A:Georgia"},
|
||
gridlineColor="F0E8D8",
|
||
plotareafill="FFFBF3", **{"plotarea.border": "E8D8B8:0.75"},
|
||
chartareafill="FDF6E8", **{"chartarea.border": "E8D8B8:0.75"},
|
||
x="14", y="0", width="13", height="18"),
|
||
# Specimen 3 · Courier New (monospace — data, telemetry, engineering)
|
||
chart("/4-Typography", title="Courier New · data mono",
|
||
**{"title.color": "1A3A1A", "title.size": "16", "title.bold": "true",
|
||
"title.font": "Courier New"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="2F8F4F",
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "2F8F4F", "axisTitle.size": "11",
|
||
"axisTitle.font": "Courier New",
|
||
"axisfont": "10:3A5A3A:Courier New"},
|
||
gridlineColor="E0EDE0",
|
||
plotareafill="F7FBF7", **{"plotarea.border": "C8DCC8:0.75"},
|
||
chartareafill="F0F7F0", **{"chartarea.border": "C8DCC8:0.75"},
|
||
x="0", y="19", width="13", height="18"),
|
||
# Specimen 4 · Verdana (friendly sans — onboarding, public-facing UI)
|
||
chart("/4-Typography", title="Verdana · friendly sans",
|
||
**{"title.color": "4A2B6A", "title.size": "16", "title.bold": "true",
|
||
"title.font": "Verdana"},
|
||
series1=f"Samples:{BELL_CSV}", binCount="18", fill="8E4DBB",
|
||
xAxisTitle="Score", yAxisTitle="Count",
|
||
**{"axisTitle.color": "8E4DBB", "axisTitle.size": "11",
|
||
"axisTitle.font": "Verdana",
|
||
"axisfont": "10:6B4A8A:Verdana"},
|
||
gridlineColor="ECE0F4",
|
||
plotareafill="FCF7FF", **{"plotarea.border": "D8C4E8:0.75"},
|
||
chartareafill="F6EDFA", **{"chartarea.border": "D8C4E8:0.75"},
|
||
x="14", y="19", width="13", height="18"),
|
||
])
|
||
|
||
# ======================================================================
|
||
# Sheet 5: "5-ML Dashboard"
|
||
#
|
||
# A cohesive six-chart "Production ML Model Report". Every chart wears the
|
||
# same corporate dashboard uniform — same typography, same frames, same
|
||
# gridlines — but each shows a different slice of the model's behavior,
|
||
# deliberately using a different color + binning strategy so the six read
|
||
# as a single dashboard at a glance.
|
||
#
|
||
# Row 1: Inference latency (ms) | Prediction confidence (%)
|
||
# Row 2: |Residual| (logit) | Token length (chars)
|
||
# Row 3: GPU utilization (%) | Cost per request ($ × 0.001)
|
||
# ======================================================================
|
||
print("\n--- 5-ML Dashboard ---")
|
||
|
||
DASH = {
|
||
"title.color": "1F2937", "title.size": "12", "title.bold": "true",
|
||
"title.font": "Helvetica Neue",
|
||
"axisTitle.color": "6B7280", "axisTitle.size": "9",
|
||
"axisTitle.font": "Helvetica Neue",
|
||
"axisfont": "8:6B7280:Helvetica Neue",
|
||
"gridlineColor": "F0F0F0",
|
||
"plotareafill": "FFFFFF", "plotarea.border": "E5E7EB:0.75",
|
||
"chartareafill": "F9FAFB", "chartarea.border": "E5E7EB:0.75",
|
||
"axisline": "9CA3AF:0.75",
|
||
"dataLabels": "false",
|
||
}
|
||
|
||
doc.batch([
|
||
add_sheet("5-ML Dashboard"),
|
||
# 1 · Inference Latency — heavy-tail, overflow-fenced, red for "watch this"
|
||
chart("/5-ML Dashboard", title="Inference Latency · p50-p99 (ms)",
|
||
series1=f"Latency:{LATENCY_CSV}", binSize="25", overflowBin="300",
|
||
fill="EF4444", **{"series.shadow": "EF4444-4-45-2-25"},
|
||
xAxisTitle="Latency (ms)", yAxisTitle="Requests", **DASH,
|
||
x="0", y="0", width="13", height="18"),
|
||
# 2 · Prediction Confidence — beta-like, axismin/max locked to 0..100
|
||
chart("/5-ML Dashboard", title="Prediction Confidence",
|
||
series1=f"Confidence:{CONFIDENCE_CSV}", binSize="5", fill="10B981",
|
||
axismin="0", majorunit="50",
|
||
xAxisTitle="Softmax confidence (%)", yAxisTitle="Samples", **DASH,
|
||
x="14", y="0", width="13", height="18"),
|
||
# 3 · Residual Magnitude — half-normal, intervalClosed=l so bin=0 catches zeros
|
||
chart("/5-ML Dashboard", title="|Residual| · model calibration",
|
||
series1=f"Residual:{ERROR_MAG_CSV}", binSize="0.25",
|
||
intervalClosed="l", fill="F59E0B",
|
||
xAxisTitle="|y - ŷ| (logit)", yAxisTitle="Samples", **DASH,
|
||
x="0", y="19", width="13", height="18"),
|
||
# 4 · Token Length — bimodal (short prompts vs long prompts)
|
||
chart("/5-ML Dashboard", title="Token Length · short vs long prompts",
|
||
series1=f"Tokens:{TOKEN_CSV}", binCount="24", fill="6366F1",
|
||
xAxisTitle="Tokens", yAxisTitle="Requests", **DASH,
|
||
x="14", y="19", width="13", height="18"),
|
||
# 5 · GPU Utilization — locked axis range so dashboard charts share scale
|
||
chart("/5-ML Dashboard", title="GPU Utilization",
|
||
series1=f"GPU:{GPU_CSV}", binSize="5", fill="8B5CF6",
|
||
axismin="0", axismax="50", majorunit="10",
|
||
xAxisTitle="Utilization (%)", yAxisTitle="Samples", **DASH,
|
||
x="0", y="38", width="13", height="18"),
|
||
# 6 · Cost per Request — log-normal, overflow-fenced, data labels with numfmt
|
||
chart("/5-ML Dashboard", title="Cost per Request ($ × 0.001)",
|
||
series1=f"Cost:{COST_CSV}", binSize="5", overflowBin="120",
|
||
fill="EC4899", dataLabels="true", **{"datalabels.numfmt": "0"},
|
||
xAxisTitle="Cost (m$)", yAxisTitle="Requests",
|
||
# DASH carries dataLabels=false; override it on this one chart.
|
||
**{k: v for k, v in DASH.items() if k != "dataLabels"},
|
||
x="14", y="38", width="13", height="18"),
|
||
])
|
||
|
||
doc.send({"command": "save"})
|
||
# context exit closes the resident, flushing the workbook to disk.
|
||
|
||
print(f"\nDone! Generated: {FILE}")
|
||
print(" 6 sheets, 29 histograms total")
|
||
print(" Sheet 0 (0-Hero): 1 magazine-grade full-bleed hero poster")
|
||
print(" Sheet 1 (1-Binning Lab): 6 charts — every binning knob, identical styling")
|
||
print(" Sheet 2 (2-Distribution Zoo): 6 canonical real-world distribution shapes")
|
||
print(" Sheet 3 (3-Theme Gallery): 6 design themes on the SAME dataset")
|
||
print(" Sheet 4 (4-Typography): 4 font-family type specimens")
|
||
print(" Sheet 5 (5-ML Dashboard): 6-chart Production ML Model Report")
|