#Data-validation
Showing 34 of 34 repositories tagged #data-validation, ranked by stars
A React component for building Web forms from JSON Schema.
Cleanlab's open-source library is the standard data-centric AI package for data quality and machine learning with messy, real-world data and labels.
Evidently is an open-source ML and LLM observability framework. Evaluate, test, and monitor any AI-powered system or data pipeline. From tabular data to Gen AI. 100+ metrics.
A light-weight, flexible, and expressive statistical data testing library
Deepchecks: Tests for Continuous Validation of ML Models & Data. Deepchecks is a holistic open-source solution for all of your AI & ML validation needs, enabling to thoroughly test your data and models from research to production.
Lightweight, extensible data validation library for Python
Data Contracts engine for the modern data stack. https://www.soda.io
Automatically find issues in image datasets and practice data-centric computer vision.
Dingo: A Comprehensive AI Data, Model and Application Quality Evaluation Tool
The data-validation toolkit for enhanced dbt (data build tool) PR review
The toolkit to test, validate, and evaluate your models and surface, curate, and prioritize the most valuable data for labeling.
Powerful CSV & Excel Import experience for SaaS 🚀 Save months building data import experience from scratch 💰
A dead simple Python string validation library.
AtroPIM is a flexible, highly configurable, modular, open-source product information management (PIM) system that extends the AtroCore data management and system integration platform.
Open Source Data Quality Monitoring.
A tool to validate data, built around Apache Spark.
pydantic --> zod data models
A simple and easy to use Data Validation library for Python.
A lightweight, declarative PySpark framework for data quality validation — check columns, rows, and entire datasets directly in your Spark pipelines
Lightweight DataFrame validation decorators for Pandas, Polars, Modin, and PyArrow. No custom types required.
⚗️ An all-in-one solution for chemical property retrieval from PubChem.
Sales insights project using Powerbi and SQL
MCP server for AI agents to analyze Excel spreadsheets through atomic operations. Like SQL for Excel. Fast, accurate, and efficient. No context overflow.
Methodology + templates for a trustworthy, traceable, self-growing AI data-analyst agent: metrics-first, independent read-only validation, and a self-growing rule library.
Never sift through endless dbt™ logs again. dbt Command Center is a free, open-source, local web application that provides a user-friendly interface to monitor and manage dbt runs.
Another library for defensive data analysis.
Stateful CSV editing MCP server for AI assistants — sessions, undo/redo, auto-save, and 39 pandas-powered tools. Works with Claude, ChatGPT, Cursor, Windsurf, Claude Code, and any MCP-compatible client. Built on FastMCP 3
A validation engine for Open Data Contract Standard (ODCS) data contracts. Define your validation rules once in a declarative YAML contract and get rich, actionable reports on your data's quality.
DCEE is a lightweight Python framework for validating data against contracts and enforcing SLA rules. Built on pandas and boto3, it provides simple, fast data validation without heavy dependencies.
:zap: Prevent downstream data quality issues by integrating the Soda Library into your CI/CD pipeline.
Python data cleaning and validation project for dengue outbreak analysis using Pandas, including missing values, duplicates, type conversion, outlier detection, and audit checks.
Declarative data quality engine. Define checks in YAML, run anywhere.
Long-form article introducing Decision Safety: a trust gate between dashboards and actions. Defines four pillars (coverage, freshness, stability, measurement risk), proposes a Decision Safety Score (0–100), shows common failure modes, and includes a copy-paste “Decision Safety Contract”+checklist to block unsafe decisions without hiding dashboards.
A long-form, practical article on data coverage: why clean dashboards still lie when datasets don’t represent the full calendar or population. Includes definitions, real failure modes (joins, filters, late data), coverage metrics, visualization patterns, anomaly/forecasting pitfalls, and reusable checklists.