sal-torch 0.1.0

Structurally Adaptive Learning — training-time sparsification for robust neural networks

sal-torch

Structurally Adaptive Learning for PyTorch

Training-time sparsification that makes neural networks structurally resilient to compression.

Install

pip install sal-torch            # core
pip install sal-torch[hf]        # + HuggingFace Trainer
pip install sal-torch[all]       # everything

from sal import SALConfig, SALCallback

config = SALConfig.auto(model)
trainer = Trainer(model=model, callbacks=[SALCallback(config)])
trainer.train()

Three lines. Any transformer. Compression-resilient.

Examples

• examples/quickstart.py — 3-line SAL training on DistilBERT
• examples/standalone_fi.py — Fragility Index scan, no training
• examples/full_control.py — manual config + standalone trainer
• examples/compare_with_without_sal.py — SAL vs. baseline under compression

New here? Start with docs/getting_started.md.

License

BSL 1.1 — free for research and evaluation. Commercial production requires a license.

Built by Cognitive Engineering in Switzerland.