tobi/grammar-inference-engine
1
0
Fork 0
Code Issues Pull requests 1 Projects Releases Packages Wiki Activity Actions

Revert "parallelize kORE outer (k, n) trials via ProcessPoolExecutor"

Browse source 
This reverts commit 0b5b0e623b.
This commit is contained in:
tobjend 2026-07-11 20:36:48 +02:00
parent 0b5b0e623b
commit ce6521ad5e
2 changed files with 18 additions and 52 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
bex/ensemble.py
View file

@ -386,18 +386,18 @@ _ALGORITHMS = {
}
def _run_kore(sequences, kmax, N, n_workers=1):
def _run_kore(sequences, kmax, N):
"""Run kOREInference, return (grammar, score) or (None, inf)."""
from .kore import kOREInference
kore = kOREInference(k_max=kmax, N=N)
result = kore.infer(sequences, n_workers=n_workers)
result = kore.infer(sequences)
if result:
_, expr, _ = result
return expr, mdl_score_simple(expr, sequences)
return None, float('inf')
def infer_ensemble(sequences, kmax=2, N=3, prefer=None, min_coverage=1.0, include_kore=False, n_workers=1):
def infer_ensemble(sequences, kmax=2, N=3, prefer=None, min_coverage=1.0, include_kore=False):
"""Run all applicable algorithms and return the best by MDL score.
Args:
@ -450,7 +450,7 @@ def infer_ensemble(sequences, kmax=2, N=3, prefer=None, min_coverage=1.0, includ
# 3. kOREInference (opt-in via include_kore=True)
if include_kore:
kore_g, kore_score = _run_kore(sequences, kmax, N, n_workers=n_workers)
kore_g, kore_score = _run_kore(sequences, kmax, N)
if kore_g:
results.append(('kOREInference', kore_g, kore_score))

62
bex/kore.py
View file

@ -13,9 +13,6 @@ Unlike the PTA→Shrink→Repair approach from Bex 2008, this follows
the journal paper (arXiv 1004.2372) exactly.
"""
import os
from concurrent.futures import ProcessPoolExecutor, as_completed
from .ikoa import ikoa
from .rwrsq import rwr_sq
from .idregex import is_deterministic
@ -56,28 +53,6 @@ def validate_k_ore(expr, k, alphabet_set=None):
return True, "OK"
def _kore_trial(args):
"""Run one (k, n) trial. Module-level for ProcessPoolExecutor.
Args:
args: (sequences, k, trial_index) — trial_index unused, for diagnostics
Returns:
(koa_automaton, expression_string, k) or None
"""
sequences, k, _ = args
G = ikoa(sequences, k, num_trials=1)
if G is None:
return None
expr = rwr_sq(G)
if expr and expr not in ('∅', 'ε'):
if is_deterministic(expr):
valid, _ = validate_k_ore(expr, k)
if valid:
return (G, expr, k)
return None
class kOREInference:
"""
|———— Algorithm 4: iDRegEx ————|
@ -97,12 +72,9 @@ class kOREInference:
self.k_max = k_max
self.N = N
def infer(self, sequences, n_workers=1):
"""Infer the best k-ORE for the given sequences.
Args:
sequences: list of token sequences
n_workers: parallel workers. >1 runs (k, n) trials concurrently.
def infer(self, sequences):
"""
Infer the best k-ORE for the given sequences.
Returns:
(koa_automaton, expression_string, best_k) or None if no valid
@ -112,25 +84,19 @@ class kOREInference:
if not sequences:
return None
trials = [(sequences, k, i)
for k in range(1, self.k_max + 1)
for i in range(self.N)]
candidates = []
if n_workers <= 1 or len(trials) < 2:
for t in trials:
result = _kore_trial(t)
if result is not None:
candidates.append(result)
else:
nw = min(n_workers, len(trials))
with ProcessPoolExecutor(max_workers=nw) as ex:
futures = [ex.submit(_kore_trial, t) for t in trials]
for f in as_completed(futures):
result = f.result()
if result is not None:
candidates.append(result)
for k in range(1, self.k_max + 1):
for _ in range(self.N):
G = ikoa(sequences, k, num_trials=1)
if G is None:
continue
expr = rwr_sq(G)
if expr and expr not in ('∅', 'ε'):
if is_deterministic(expr):
valid, _ = validate_k_ore(expr, k)
if valid:
candidates.append((G, expr, k))
if not candidates:
return None