mirror of
https://github.com/luxfi/torus-ml.git
synced 2026-10-30 15:15:23 +00:00
zeekay
cf51efefb9
A version written into a manifest is written into every manifest that wants it, which is how eslint came to exist in a hundred and four forms across this estate and how a package came to resolve twice at run time. The catalog moves the number out of the manifest: a manifest names the package, the workspace names the version. Four are held below their latest because the newer release cannot be used, and the reason is recorded rather than the version alone: @noble/hashes 2 renamed every subpath the web3 stack imports, @splinetool/runtime 2 depends on a package that was never published, @dynamic-labs 5.4.1 the same, and react-resizable-panels 3 points its types condition at the bundle.
Torus ML
Machine Learning framework for encrypted data using Fully Homomorphic Encryption.
Overview
Torus ML enables training and inference on encrypted data without decryption. Part of the Lux FHE ecosystem.
Features
- Privacy-preserving ML: Train models on encrypted data
- Quantization-aware training: Optimized for FHE computation
- scikit-learn compatible: Familiar API for ML practitioners
- GPU acceleration: native CUDA / Metal / WebGPU via the Torus compiler over the luxfi/gpu runtime
Installation
pip install torus-ml
Or from source:
git clone https://github.com/luxfi/torus-ml
cd torus-ml
pip install -e .
Quick Start
from torus.ml import FHEModelClient, FHEModelServer
from sklearn.datasets import make_classification
# Train a model
X, y = make_classification(n_samples=1000)
model = FHEModelClient()
model.fit(X, y)
# Compile for FHE
model.compile(X)
# Encrypted inference
encrypted_X = model.encrypt(X[:1])
encrypted_pred = model.predict(encrypted_X)
pred = model.decrypt(encrypted_pred)
Integration
- luxfi/fhe - Core FHE library (Go)
- luxfi/torus - Python FHE framework
- luxfi/fhe-compiler - LLVM compiler
- luxfi/lattice - Lattice primitives
License
BSD 3-Clause Clear License