pandas-plots 1.4.6

A collection of helper for table handling and visualization

pandas-plots

a comprehensive python package for enhanced data visualization and analysis with pandas dataframes. provides a high-level api for creating beautiful tables, plots, venn diagrams, and utility functions with minimal code.

• pandas-plots
  • 🚀 features
  • 📦 installation
  • 🛠️ usage
  • examples
    • styled table
    • table description
    • upset plot
    • uml graph
    • set filter
    • sankey diagram
    • box plot with violin overlay
    • box plot with statistics
    • venn diagrams
  • 📚 api reference
    • table utilities tbl
    • plotting functions pls
    • venn diagrams ven
    • helper functions hlp
  • ⚙️ configuration
    • environment settings
  • 🧩 prerequisites
  • 🤝 contributing
  • 📄 license
  • 🏷️ tags

🚀 features

• table utilities (tbl): enhanced table display and description functions
• plotting functions (pls): comprehensive visualization tools with plotly
• venn diagrams (ven): easy-to-create venn diagrams for set analysis
• helper functions (hlp): utility functions for common data operations
• constants (const): useful constants like color palettes

📦 installation

# using uv (recommended)
uv add -U pandas-plots

🚨 this package relies on installed ungoogled-chromium to make static image conversion work. regular chromium is deprecated

🛠️ usage

from pandas_plots import tbl, pls, ven, hlp, const

# for public examples: load sample dataset from seaborn
import seaborn as sb
df = sb.load_dataset('taxis')

examples

styled table

tbl.pivot_df(
    df[["color", "payment", "fare"]],
    total_mode="sum",
    total_axis="xy",
    data_bar_axis=None,
    total_exclude=True,
    pct_axis="xy",
    precision=0,
    heatmap_axis="xy",
    kpi_mode="rag_abs",
    kpi_rag_list=[1000, 10000],
    swap=True,
    font_size_td=12,
    font_size_th=14,
)

table description

tbl.describe_df(
    df,
    caption="taxis",
    top_n_uniques=10,
    top_n_chars_in_columns=10,
    top_n_chars_in_index=15,
)

upset plot

pls.plot_upset(
    df_upset,
    include_false_subsets=False,
    orientation="horizontal",
)

uml graph

metrics = pls.plot_uml_graph()

set filter

filter = hlp.get_duckdb_filter_n(
    con,
    "from Tumor",
    FILTERS,
    # distinct_metric="z_pat_id",
)

counts: rows
---
n = 3_241_401                                     (100.0%) ██████████████████████████████
└ [2020-2023.07]:                   n = 2_633_644  (81.3%) ░░░░░░████████████████████████
└ [not z_is_dco]:                   n = 2_547_636  (78.6%) ░░░░░░░███████████████████████
└ [keine M1]:                       n = 2_305_215  (71.1%) ░░░░░░░░░█████████████████████
└ [keine Verstorbenen < 180 Tage]:  n = 2_132_064  (65.8%) ░░░░░░░░░░░███████████████████
└ [lympho- und mesoendokr. Tumore]:    n = 27_653   (0.9%) ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░

sankey diagram

_ = pls.plot_sankey(width=1000)

box plot with violin overlay

_ = pls.plot_box(df['fare'], height=400, violin=true)

box plot with statistics

_ = pls.plot_boxes_large(df[["dropoff_borough","distance"]])

venn diagrams

# show venn diagram for 3 sets
set_a = {'ford','ferrari','mercedes', 'bmw'}
set_b = {'opel','bmw','bentley','audi'}
set_c = {'ferrari','bmw','chrysler','renault','peugeot','fiat'}

_df, _details = ven.show_venn3(
    title="taxis",
    a_set=set_a,
    a_label="cars1",
    b_set=set_b,
    b_label="cars2",
    c_set=set_c,
    c_label="cars3",
    verbose=0,
    size=8,
)

📚 api reference

table utilities (tbl)

function	description
show_num_df()	displays a table as styled version with additional information
describe_df()	alternative version of pandas describe() function
descr_db()	short description for a duckdb relation
pivot_df()	gets a pivot table of a 3 column dataframe (or 2 columns if no weights are given)
print_summary()	shows statistics for a pandas dataframe or series

plotting functions (pls)

function	description
plot_box()	auto annotated boxplot w/ violin option
plot_boxes()	multiple boxplots (annotation is experimental)
plot_stacked_bars()	shortcut to stacked bars
plot_bars()	standardized bar plot for categorical column with confidence intervals
plot_histogram()	histogram for one or more numerical columns
plot_joints()	joint plot for exactly two numerical columns
plot_quadrants()	quickly shows a 2x2 heatmap
plot_facet_stacked_bars()	stacked bars for a facet value as subplots
plot_sankey()	generates a sankey diagram
plot_pie()	generates a pie chart
plot_box_large()	for large datasets using seaborn
plot_boxes_large()	for large datasets using seaborn
plot_histogram_large()	for large datasets using seaborn
plot_upset()	generates an upset plot based on upsetplot
plot_uml_graph()	generates a uml graph based on mermaid for structured data

venn diagrams (ven)

function	description
show_venn2()	displays a venn diagram for 2 sets
show_venn3()	displays a venn diagram for 3 sets

helper functions (hlp)

function	description
to_series()	converts a dataframe to a series
mean_confidence_interval()	calculates mean and confidence interval for a series
wrap_text()	formats strings or lists to a given width
replace_delimiter_outside_quotes()	replaces delimiters only outside of quotes in csv imports
create_barcode_from_url()	creates a barcode from a given url
add_datetime_col()	adds a datetime column to a dataframe (chainable)
show_package_version()	prints version of a list of packages
get_os()	helps identify and ensure operating system at runtime
add_bitmask_label()	adds a column that resolves a bitmask column into human-readable labels
find_cols()	finds all columns in a list that contain any of the given stubs
add_measures_to_pyg_config()	adds measures to a pygwalker config file
get_tum_details()	prints details of a specific tumor (requires connection to clinical cancer data)
get_duckdb_filter_n()	print rowcounts for cascading filters in duckdb with ansi bars
print_filter()	print filter as markdown sql codeblock
is_ipynb()	detects if code is running in jupyter notebook
prepend_uv_header()	prepends uv header to a .py script to make it executable for uv run command
create_py_script()	creates a .py script from a .ipynb file
setup_rendering()	triggers clean(er) rendering of plots and pandas tables to markdown

⚙️ configuration

environment settings

# * set theme: light / dark
# * this will affect all plots
os.environ['THEME'] = 'light'

# * set renderer: svg / png for static images, '' for interactive plots
# * note: only static images will be rendered in markdown
os.environ['RENDERER'] = 'svg'

🧩 prerequisites

• python 3.10+: compatible with python versions 3.10 - 3.13
• uv: uv is recommended for package management

⚠️ this package depends on numpy<2.0.0 since UpSetPlot is still tied to the previous versions

🤝 contributing

contributions are welcome! please feel free to submit a pull request. for major changes, please open an issue first to discuss what you would like to change.

📄 license

this project is licensed under the MIT license - see the license file for details.

🏷️ tags

#pandas #visualizations #statistics #data-science #data-analysis #python