83 lines
3.1 KiB
Python
83 lines
3.1 KiB
Python
from __future__ import annotations
|
|
|
|
import numpy as np
|
|
|
|
from algebra.cga import cga_inner
|
|
from algebra.holonomy import holonomy_encode, holonomy_similarity
|
|
from language_packs import load_pack
|
|
from language_packs.compiler import compile_entries_to_manifold, load_pack_entries
|
|
from morphology.registry import load_morphology
|
|
|
|
|
|
def _encode(manifold, tokens: list[str]) -> np.ndarray:
|
|
return holonomy_encode([manifold.get_versor(t) for t in tokens])
|
|
|
|
|
|
def test_aligned_clauses_have_higher_similarity_than_unrelated():
|
|
_, en = load_pack("en_minimal_v1")
|
|
_, he = load_pack("he_logos_micro_v1")
|
|
_, grc = load_pack("grc_logos_micro_v1")
|
|
|
|
en_clause = _encode(en, ["word", "beginning", "with", "truth"])
|
|
he_clause = _encode(he, ["דבר", "ראשית", "אמת"])
|
|
grc_clause = _encode(grc, ["λόγος", "ἀρχή", "ἀλήθεια"])
|
|
grc_unrelated = _encode(grc, ["λόγος", "ἀρχή", "ζωή"])
|
|
|
|
aligned = (np.linalg.norm(en_clause - he_clause) + np.linalg.norm(en_clause - grc_clause)) / 2.0
|
|
unrelated = np.linalg.norm(en_clause - grc_unrelated)
|
|
assert aligned < unrelated
|
|
|
|
|
|
def test_triple_alignment_closer_than_other_triples():
|
|
_, en = load_pack("en_minimal_v1")
|
|
_, he = load_pack("he_logos_micro_v1")
|
|
_, grc = load_pack("grc_logos_micro_v1")
|
|
|
|
word_trip = [
|
|
en.get_versor("word"),
|
|
he.get_versor("דבר"),
|
|
grc.get_versor("λόγος"),
|
|
]
|
|
aligned_score = np.mean(
|
|
[
|
|
cga_inner(en.get_versor("word"), he.get_versor("דבר")),
|
|
cga_inner(en.get_versor("word"), grc.get_versor("λόγος")),
|
|
cga_inner(he.get_versor("דבר"), grc.get_versor("λόγος")),
|
|
]
|
|
)
|
|
misaligned_score = np.mean(
|
|
[
|
|
cga_inner(en.get_versor("word"), he.get_versor("ראשית")),
|
|
cga_inner(en.get_versor("word"), grc.get_versor("πνεῦμα")),
|
|
cga_inner(he.get_versor("דבר"), grc.get_versor("ἀρχή")),
|
|
]
|
|
)
|
|
assert aligned_score > misaligned_score
|
|
|
|
|
|
def test_word_order_permutation_changes_holonomy():
|
|
_, en = load_pack("en_minimal_v1")
|
|
a = _encode(en, ["word", "truth", "light", "life"])
|
|
b = _encode(en, ["life", "word", "truth", "light"])
|
|
assert abs(holonomy_similarity(a, b) - holonomy_similarity(a, a)) > 1e-4
|
|
|
|
|
|
def test_same_root_hebrew_forms_land_closer_than_unrelated_noun():
|
|
_, he = load_pack("he_logos_micro_v1")
|
|
|
|
singular = he.get_versor("דבר")
|
|
plural = he.get_versor("דברים")
|
|
unrelated = he.get_versor("ראשית")
|
|
|
|
assert cga_inner(singular, plural) > cga_inner(singular, unrelated)
|
|
|
|
|
|
def test_structured_morphology_improves_same_root_hebrew_resonance():
|
|
entries = load_pack_entries("he_logos_micro_v1")
|
|
tag_only = compile_entries_to_manifold(entries)
|
|
structured = compile_entries_to_manifold(entries, load_morphology("he_logos_micro_v1"))
|
|
|
|
tag_only_score = cga_inner(tag_only.get_versor("דבר"), tag_only.get_versor("דברים"))
|
|
structured_score = cga_inner(structured.get_versor("דבר"), structured.get_versor("דברים"))
|
|
|
|
assert structured_score > tag_only_score
|