Install

pip install analysistoolbox

Requires Python 3.9 or later.

Configure plot style

Add this to the top of any analysis notebook to get clean, consistent charts:

import seaborn as sns
sns.set(style="white", font="Arial", context="paper")

Your first analysis

from analysistoolbox.data_processing import CreateDataOverview
import pandas as pd

df = pd.read_csv('your_data.csv')
CreateDataOverview(dataframe=df, plot_missingness=True)

CreateDataOverview returns a summary with data types, missing value counts, and distribution plots for every column — one function call instead of twenty.

Dependencies

Analysis Tool Box installs these automatically:

• numpy, pandas, scipy, scikit-learn — core data stack
• matplotlib, seaborn — visualization
• statsmodels, xgboost — modeling
• pymetalog — Metalog distribution fitting
• fuzzywuzzy, Levenshtein — entity matching
• sympy — symbolic math (Calculus module)
• geopandas, folium — geospatial (Geospatial module)
• tableone — clinical summary tables

Next steps

Browse the module reference or jump straight to the Simulations and Probability modules for the intelligence analysis use cases.