Giorgio Gilestro
840b6b00b3
Re-scopes Layer 2 into a cheaper, architecture-general neural collapse proof
before the LLM rung. Realises the same Wright–Fisher abstractions in real trained
generative models on a fully-synthetic sandbox with an exact oracle, reusing
knowledge.metrics/truth/seeding and the output contract so neural curves overlay
the Layer-1 analytic curves.
- src/neural/: synthetic token-grammar sandbox (lossless identity + stochastic
style), ExactOracle, HistogramModel bridge, generation loop, experiment runner
- HARD GATE passed: histogram lineage reproduces Layer 1 exactly (neutral decay,
exact H_eq, tracks run_lineage) — tests/test_neural_validation.py
- torch models: autoregressive RNN + MLP (VAE implemented, not yet fidelity-
passing); determinism seeding derived from the SeedSequence stream
- N0 bridge (neural g*=0.047 ≈ Layer-1 0.048), N1 collapse-in-weights, N2 phase
boundary, N5 architecture-generality (collapse + grounding-rescue in histogram
+ RNN + MLP). Manifests/configs committed; parquet gitignored, hashes tracked
- additive backward-compatible save_artifacts extension; Makefile neural targets
Finding: neural smoothing partially resists H-collapse, so forward-KL and tail
survival are the sharp neural collapse metrics (H is smooth, per Layer 1).
92 tests green. Remaining (tasks/todo.md): N4 merge, N2 refine, N3/N6, VAE
fidelity, MNIST tier, figures. LLM/LoRA rung and C3 deferred.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
38 lines
1.2 KiB
TOML
38 lines
1.2 KiB
TOML
[project]
|
|
name = "lamarckian-society"
|
|
version = "0.1.0"
|
|
description = "Layer 1 analytical core: knowledge transmission as a Wright-Fisher process."
|
|
readme = "README.md"
|
|
requires-python = ">=3.11"
|
|
dependencies = [
|
|
"numpy>=1.26",
|
|
"scipy>=1.11",
|
|
"pandas>=2.1",
|
|
"pyarrow>=15",
|
|
"matplotlib>=3.8",
|
|
"pydantic>=2.5",
|
|
"pyyaml>=6.0",
|
|
]
|
|
|
|
[project.optional-dependencies]
|
|
dev = ["pytest>=8.0"]
|
|
# Layer 1.5 neural existence proof. Torch is only needed from Stage C (RNN/VAE/MLP);
|
|
# Stages A-B (synthetic sandbox + histogram bridge) are pure NumPy and run in the base env.
|
|
# The default PyPI torch wheel is CUDA-enabled (cu13, matching the RTX A4000 driver).
|
|
# Install with `uv sync --extra neural`.
|
|
neural = ["torch>=2.2"]
|
|
# The real-MNIST secondary-confirmation tier only. Install with `uv sync --extra mnist`.
|
|
mnist = ["torchvision>=0.17"]
|
|
|
|
[build-system]
|
|
requires = ["hatchling"]
|
|
build-backend = "hatchling.build"
|
|
|
|
# src-layout: src/knowledge/ is importable as `knowledge` (the normative package
|
|
# name the scientific-validation conformance tests import). src/neural/ is Layer 1.5.
|
|
[tool.hatch.build.targets.wheel]
|
|
packages = ["src/knowledge", "src/neural"]
|
|
|
|
[tool.pytest.ini_options]
|
|
testpaths = ["tests"]
|
|
addopts = "-q"
|