Skip to main content

An all-encompassing, automated toolkit for deep Exploratory Data Analysis and reporting.

Project description

Decyphr

An all-encompassing, automated toolkit for generating deep, professional, and interactive Exploratory Data Analysis (EDA) reports with a single line of code.

Decyphr is designed to accelerate the data science workflow by automating the tedious and time-consuming process of initial data exploration. It goes beyond basic profiling to provide deep statistical insights, advanced machine learning-driven analysis, and a stunning, presentation-ready report that is as beautiful as it is informative.

(Note: You will replace this with a screenshot of your final report)

Key Features

Decyphr provides a comprehensive suite of analyses, intelligently triggered based on your data's characteristics:

Complete Overview: Instant summary of dataset shape, memory usage, variable types, and data quality warnings.

Deep Univariate Analysis: Detailed statistical profiles, histograms, and frequency charts for every variable.

Multivariate Analysis: Stunning, interactive heatmaps for both linear (Pearson) and non-linear (Phik) correlations.

Advanced Data Quality: Automatically detects constant columns, whitespace issues, and potential outliers using multiple methods (IQR, Isolation Forest).

Statistical Inference: Performs automated Hypothesis Testing (T-Tests, ANOVA, Chi-Squared) to uncover statistically significant relationships.

Machine Learning Insights: PCA: Analyzes dimensionality reduction possibilities. Clustering: Automatically finds hidden segments in your data using K-Means. Feature Importance: Trains a baseline model to identify the most predictive features when a target is provided. Explainable AI (XAI): Generates SHAP summary plots to explain how your features impact model predictions.

Specialized Analysis: Includes dedicated modules for Deep Text Analysis (Sentiment, NER, Topics), Time-Series Decomposition, and Geospatial Mapping.

Data Drift Detection: Compare two datasets to quantify changes in data distribution over time.

High-End Interactive Report: A beautiful, modern dashboard with a toggleable light/dark theme, responsive charts, and a professional UI/UX.

Quick Start

  1. Installation

Install Decyphr and its core dependencies from the root of the project directory:

Navigate to the decyphr_project folder

pip install -e .

To enable all features, including deep text and geospatial analysis, install the [all] optional dependencies:

pip install -e ".[all]"

For deep text analysis, you will also need to download the SpaCy language model:

python -m spacy download en_core_web_sm

  1. Generate Your First Report

Create a Python script (analyze.py) and add the following code. Just point it to your dataset and let Decyphr do the rest.

analyze.py

import decyphr

Generate a comprehensive report for a single dataset

decyphr.analyze(filepath="path/to/your/data.csv")

--- OR ---

Generate a report that includes target-driven analysis

decyphr.analyze(

filepath="path/to/your/training_data.csv",

target="column_to_predict"

)

Run the script from your terminal:

python analyze.py

Your professional, interactive HTML report will be automatically saved in a new decyphr_reports/ directory.

Project Structure

This project uses a highly modular "plugin" architecture to ensure it is robust, maintainable, and easy to extend. All analysis and visualization logic is separated into self-contained modules located in the src/decyphr/analysis_plugins/ directory.

Contributing

Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated. Please feel free to fork the repo and create a pull request, or open an issue with suggestions.

License

Distributed under the MIT License. See LICENSE file for more information.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

decyphr-1.0.0.tar.gz (50.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

decyphr-1.0.0-py3-none-any.whl (84.4 kB view details)

Uploaded Python 3

File details

Details for the file decyphr-1.0.0.tar.gz.

File metadata

  • Download URL: decyphr-1.0.0.tar.gz
  • Upload date:
  • Size: 50.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.7

File hashes

Hashes for decyphr-1.0.0.tar.gz
Algorithm Hash digest
SHA256 6e226d592cb5e54d1b7a764746eec26c428fe2023a3fb459500ed398ed30cd31
MD5 751b9fa2691d07a239efde5ce30248d3
BLAKE2b-256 7f932eb35b1a02b76aef68575dd3d0986b137ecc870f67e9420fddaba28827a1

See more details on using hashes here.

File details

Details for the file decyphr-1.0.0-py3-none-any.whl.

File metadata

  • Download URL: decyphr-1.0.0-py3-none-any.whl
  • Upload date:
  • Size: 84.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.7

File hashes

Hashes for decyphr-1.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 e02c54204f798a67647033c5ddc49ce053a029f2219f83127c48e12b9778a18a
MD5 6e3581a0cb6dc9e6cc2354620232f498
BLAKE2b-256 1ade5e01f6514546ec280bc5ef89daa65909c4e54ce13ed994032d9e5e26e22f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page