Metadata-Version: 2.4
Name: deepdrift
Version: 0.4.0
Summary: Production-grade Kinetic Monitoring for Neural Networks (Semantic Velocity)
Home-page: https://github.com/Eutonics/DeepDrift
Author: Alexey Evtushenko
Author-email: alexey@eutonics.ru
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Description-Content-Type: text/plain
Requires-Dist: torch>=1.10.0
Requires-Dist: numpy>=1.21.0
Requires-Dist: scikit-learn>=1.0.0
Requires-Dist: pandas>=1.3.0
Requires-Dist: tqdm
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

DeepDrift treats neural networks as dynamic systems, measuring the 'velocity' of hidden states to detect OOD, Fraud, and Hallucinations. Version 0.4.0 introduces Sparse Sampling (<1% overhead) and Robust Thresholding.
