Training Data Python Packages

amazoncaptcha

Pure Python, lightweight, Pillow-based solver for Amazon's text captcha.

snorkel

A system for quickly generating training data with weak supervision

augraphy

Augmentation pipeline for rendering synthetic paper printing, faxing, scanning and copy machine processes

dtflow

Inspect, clean, transform and ship LLM training data from one CLI; dt view is the interactive entry point.

composeml

A machine learning tool for automated prediction engineering. It allows you to easily structure prediction problems and generate labels for supervised learning.

ydata-synthetic

Synthetic data generators for tabular and time-series data

sediment-core

Self-hosted evidence store for coding agents.

crovia-tacet

TACET — verifiable silence for AI training-data disclosure. Hourly epochs opened by drand, closed by Bitcoin; a public predicate over model cards; offline-verifiable proofs of absence as crovia.seal.v1. Spec, reference, conformance vectors, live operator.

sediment-derive

Self-hosted evidence store for coding agents.

sediment-export

Self-hosted evidence store for coding agents.

sediment-capture

Self-hosted evidence store for coding agents.

sediment-api

Self-hosted evidence store for coding agents.

sediment-cli

Self-hosted evidence store for coding agents.

pdf-anonymizer-core

CLI, Web API, and an SDK anonymizing large PDF files, Excel, Word docs, CSVs

pdf-anonymizer-cli

CLI, Web API, and an SDK anonymizing large PDF files, Excel, Word docs, CSVs

gsj-harness-rollout-server

Rollout server for RL over a temporally-scoped corpus: (case, timestep, prompt) → a pinned agent in a sandbox truncated at that timestep → one validated, training-ready trajectory.

pdf-anonymizer-api

CLI, Web API, and an SDK anonymizing large PDF files, Excel, Word docs, CSVs

axiom-all

An open-source platform for building, fine-tuning, evaluating, and deploying AI models.

augmenty

Augmenty is an augmentation library based on spaCy for augmenting texts.

dropoutt

Pre-flight checks for LLM training data. Scan, fingerprint, and map corpora before you burn a training run.

id-extract

CLI, Web API, and an SDK anonymizing large PDF files, Excel, Word docs, CSVs

ladon-crawl

A Python framework for building structured, resumable web crawlers — designed for domains where data quality matters.

skweak

skweak: A software toolkit for weak supervision applied to NLP tasks

s2s-certify

Physics-audit engine for Physical AI. 16 biomechanical laws + triple coherence firewall — catches synthetic and corrupted sensor data before model training. pip install s2s-certify