deepanwadhwa/zink
A Python package for zero-shot text anonymization using Transformer-based NER models.
This tool helps data privacy officers, compliance managers, and data analysts protect sensitive personal information in text documents. It takes raw text containing details like names, locations, or company names, and either redacts these details (e.g., replacing 'John Doe' with '[PERSON_REDACTED]') or replaces them with synthetic, realistic-looking data. This ensures privacy while still allowing the text to be used for analysis or other purposes.
Available on PyPI.
Use this if you need to quickly and accurately remove or replace sensitive identifying information from large volumes of text without needing pre-labeled examples for every type of data you want to protect.
Not ideal if you require extremely high precision for highly specialized or obscure entity types where a custom, fine-tuned model would offer better performance.
Stars
82
Forks
2
Language
Python
License
Apache-2.0
Category
Last pushed
Dec 16, 2025
Commits (30d)
0
Dependencies
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/deepanwadhwa/zink"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
tensorflow/privacy
Library for training machine learning models with privacy for training data
meta-pytorch/opacus
Training PyTorch models with differential privacy
tf-encrypted/tf-encrypted
A Framework for Encrypted Machine Learning in TensorFlow
awslabs/fast-differential-privacy
Fast, memory-efficient, scalable optimization of deep learning with differential privacy
privacytrustlab/ml_privacy_meter
Privacy Meter: An open-source library to audit data privacy in statistical and machine learning...