dataclocklib 0.1.4

Data clock charts using matplotlib.

Data Clock Visualisation Library

[!NOTE] This library is a work in progress and is frequently updated.

TODO

• Publish to PyPI.
• Add unit tests for chart generation & data aggregation.
• Add tutorial section to GitHub Page documentation.

Introduction

This library allows the user to create data clock graphs, using the matplotlib Python library.

Data clocks visually summarise temporal data in two dimensions, revealing seasonal or cyclical patterns and trends over time. A data clock is a circular chart that divides a larger unit of time into rings and subdivides it by a smaller unit of time into wedges, creating a set of temporal bins.

These temporal bins are symbolised using graduated colors that correspond to a count or aggregated value taking place in each time period.

The table below details the currently supported chart modes and the corresponding rings and wedges:

Mode	Rings	Wedges	Description
YEAR_MONTH	Years	Months	Years / January - December.
YEAR_WEEK	Years	Weeks	Years / weeks 1 - 52.
WEEK_DAY	Weeks	Days of the week	Weeks 1 - 52 / Monday - Sunday.
DOW_HOUR	Days of the week	Hour of day	Monday - Sunday / 24 hours.
DAY_HOUR	Days	Hour of day	Days 1 - 356 / 24 hours.

The full documentation can be viewed on the project GitHub Page.

Example charts

Chart examples have been generated using UK Department for Transport data 2010 - 2015.

import pandas as pd
from dataclocklib.charts import dataclock

data = pd.read_parquet(
    "https://raw.githubusercontent.com/andyrids/dataclocklib/main/tests/data/traffic_data.parquet.gzip"
)

graph_data, fig, ax = dataclock(
    data=data.query("Date_Time.dt.year.ge(2015)"),
    date_column="Date_Time",
    agg_column="Number_of_Casualties",
    agg="sum",
    mode="DOW_HOUR",
    cmap_name="CMRmap_r",
    chart_title="UK Car Accident Casualties 2015",
    chart_subtitle=None,
    chart_source="www.kaggle.com/datasets/silicon99/dft-accident-data"
)

import pandas as pd
from dataclocklib.charts import dataclock

data = pd.read_parquet(
    "https://raw.githubusercontent.com/andyrids/dataclocklib/main/tests/data/traffic_data.parquet.gzip"
)

graph_data, fig, ax = dataclock(
    data=data.query("Date_Time.dt.year.eq(2010)"),
    date_column="Date_Time",
    agg_column=None,
    agg="count",
    mode="DOW_HOUR",
    cmap_name="RdYlGn_r",
    chart_title="UK Car Accidents 2010",
    chart_subtitle=None,
    chart_source="www.kaggle.com/datasets/silicon99/dft-accident-data"
)

Installation

You can install using pip:

python -m pip install dataclocklib

Development Installation

Astral uv is used as the Python package manager. To install uv see the installation guide @ uv documentation.

Clone the repository:

git clone [e-mail]:andyrids/dataclocklib.git
cd dataclocklib

Sync the dependencies, including the dev dependency group and optional dependencies with uv:

uv sync --all-extras

Activate the virtual environment:

. .venv/bin/activate

Sphinx documentation

cd docs
make html

Dependencies

dataclocklib
├── matplotlib v3.10.0
├── pandas[parquet] v2.2.3
├── sphinx v8.1.3 (extra: docs)
├── sphinx-autobuild v2024.10.3 (extra: docs)
└── sphinx-rtd-theme v3.0.2 (extra: docs)