ZacharyZhang-NY
e81923f62e
The 300-probe sample this harness scored by modality read the screenshot modality 0.818 where all 320 of its answerable queries read 0.659, because striding a file-ordered list and truncating at 300 reshuffles the modality mix. Recall is now scored over every answerable query (2,928) and latency over a stratified sample, which is what latency needs and what recall does not. The separate conversation sub-pass is gone: conversation is inside the suite, and by_category already carries its six LongMemEval types.
369 lines
16 KiB
Python
369 lines
16 KiB
Python
"""What each retrieval leg is actually worth, on the frozen suite, asked as its holder.
|
|
|
|
The product fuses three legs with reciprocal rank fusion: a sparse page, a dense vector leg, and a
|
|
picture leg that only speaks up when the words come back thin. Each one costs latency and a model
|
|
snapshot, and the suite is the only honest referee. This harness turns each leg on and off over
|
|
**every answerable query in the suite** and prints recall, first-hit quality, latency, and
|
|
per-modality recall in one table, so a default like `dense_embeddings_enabled=False` is a
|
|
measurement rather than a preference. Recall is scored whole because a 300-query sample reads a
|
|
modality a few points off; latency is timed on a stratified sample because that is what a sample
|
|
is for.
|
|
|
|
Run: python -m bench.leg_check [--probes 300] # --probes sizes the timing sample only
|
|
Writes: bench/reports/LEGS.md, bench/reports/leg_check.json
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import argparse
|
|
import json
|
|
import shutil
|
|
import statistics
|
|
import sys
|
|
import time
|
|
from pathlib import Path
|
|
from typing import Any
|
|
|
|
from bench.leg_report import write_markdown
|
|
from bench.metrics import aggregate, newest_gold_first, score_query
|
|
from bench.query_set import load_queries, scale_probes, source_times
|
|
from bench.report import git_commit
|
|
from bench.run import latency
|
|
from bench.scratch import copy_store
|
|
from bench.systems import machine_facts
|
|
|
|
DEFAULT_HOME = Path("bench/runs/dense_scale")
|
|
DEFAULT_SCRATCH = Path("bench/runs/leg_check")
|
|
DEFAULT_OBJECTS = Path("bench/runs/baseline/objects")
|
|
DEFAULT_CLIP_INDEX = Path("bench/runs/leg_check_index")
|
|
DEFAULT_CLIP_MODEL = Path("bench/models/chinese-clip-vit-base-patch16")
|
|
DEFAULT_EMBEDDING_MODEL = Path("bench/models/paraphrase-multilingual-MiniLM-L12-v2")
|
|
ENCODE_SAMPLE = 200
|
|
|
|
# The page the product asks for by default; measure() checks it against Settings, so "as shipped" cannot rot into a hypothesis.
|
|
SHIPPED_TOP_K = 24
|
|
|
|
# (label, atoms asked, dense leg, picture leg); the picture leg answers only on a thin page, which is the product's rule.
|
|
CONFIGS: list[dict[str, Any]] = [
|
|
{"label": "words only", "top_k": 10, "dense": False, "visual": False},
|
|
{"label": "words only (the configured 24)", "top_k": SHIPPED_TOP_K, "dense": False, "visual": False},
|
|
{"label": "words only, wider page", "top_k": 30, "dense": False, "visual": False},
|
|
{"label": "+ dense leg", "top_k": 10, "dense": True, "visual": False},
|
|
{"label": "+ dense leg, wider page", "top_k": 30, "dense": True, "visual": False},
|
|
{"label": "+ picture leg", "top_k": 10, "dense": False, "visual": True},
|
|
{"label": "+ picture leg, wider page", "top_k": 30, "dense": False, "visual": True},
|
|
]
|
|
|
|
|
|
def prepare(
|
|
home: Path,
|
|
scratch: Path,
|
|
objects: Path,
|
|
clip_index: Path,
|
|
*,
|
|
dense: bool,
|
|
visual: bool,
|
|
clip_model: Path,
|
|
embedding_model: Path,
|
|
) -> Path:
|
|
"""Build a scratch store that can run every leg: vectors, the picture index, and the bytes.
|
|
|
|
The picture index and the CLIP snapshot are 48 minutes of encoding, so they are carried in from
|
|
wherever they were built rather than rebuilt per row.
|
|
"""
|
|
copy_home, _ = copy_store(home, scratch)
|
|
if visual:
|
|
for name in ("clip_image_index.npy", "clip_image_index_atoms.json"):
|
|
source = clip_index / name
|
|
if not source.exists():
|
|
raise SystemExit(
|
|
f"{source}: the picture index has not been built; run `ely-eye index-visual` "
|
|
"with ELY_EYE_CLIP_MODEL_DIR set, or point --clip-index at its cache"
|
|
)
|
|
(copy_home / "cache").mkdir(parents=True, exist_ok=True)
|
|
shutil.copy2(source, copy_home / "cache" / name)
|
|
if not (clip_model / "onnx" / "model_fp16.onnx").exists():
|
|
raise SystemExit(f"{clip_model / 'onnx' / 'model_fp16.onnx'}: the CLIP snapshot is missing")
|
|
if not objects.is_dir():
|
|
raise SystemExit(f"{objects}: the picture leg needs the object store to check image bytes")
|
|
link = copy_home / "objects"
|
|
if not link.exists() and not link.is_symlink():
|
|
link.symlink_to(objects.resolve())
|
|
if dense and not (embedding_model / "config.json").exists():
|
|
raise SystemExit(f"{embedding_model}: the embedding snapshot is missing; the dense leg cannot run")
|
|
|
|
return copy_home
|
|
|
|
|
|
def _dense_vectors(store: Path) -> int:
|
|
"""Vectors the source store already carries, read without opening it for write."""
|
|
import sqlite3
|
|
|
|
import sqlite_vec
|
|
from ely_eye import vector_index
|
|
|
|
if vector_index.TABLE != "vec_atoms":
|
|
raise SystemExit(f"the dense table is now {vector_index.TABLE}; update this harness's count")
|
|
if not store.is_file():
|
|
raise SystemExit(f"{store}: no product database")
|
|
with sqlite3.connect(f"file:{store}?mode=ro", uri=True) as connection:
|
|
connection.enable_load_extension(True)
|
|
sqlite_vec.load(connection)
|
|
connection.enable_load_extension(False)
|
|
if not connection.execute("SELECT 1 FROM sqlite_master WHERE name = 'vec_atoms'").fetchone():
|
|
return 0
|
|
return int(connection.execute("SELECT count(*) FROM vec_atoms").fetchone()[0])
|
|
|
|
|
|
def picture_facts(clip_index: Path) -> dict[str, Any]:
|
|
"""What the picture index really holds, read off the artifact instead of remembered in prose."""
|
|
import numpy as np
|
|
|
|
vectors_path = clip_index / "clip_image_index.npy"
|
|
atoms_path = clip_index / "clip_image_index_atoms.json"
|
|
vectors = np.load(vectors_path)
|
|
with atoms_path.open(encoding="utf-8") as handle:
|
|
try:
|
|
atoms = json.load(handle)
|
|
except json.JSONDecodeError as error:
|
|
raise ValueError(f"{atoms_path}: the picture index's atom list is unreadable ({error})") from error
|
|
if len(atoms) != int(vectors.shape[0]):
|
|
raise SystemExit(
|
|
f"{clip_index}: the picture index disagrees with itself — {len(atoms)} atom ids for "
|
|
f"{vectors.shape[0]} vectors"
|
|
)
|
|
return {
|
|
"path": str(clip_index),
|
|
"atoms": len(atoms),
|
|
"dim": int(vectors.shape[-1]),
|
|
"bytes": int(vectors_path.stat().st_size),
|
|
}
|
|
|
|
|
|
def encode_rate(clip_model: Path, objects: Path, count: int) -> dict[str, Any]:
|
|
"""Pictures per second the encoder manages, measured on a sample instead of remembered."""
|
|
import time
|
|
|
|
from ely_eye.clip_encoder import ClipEncoder
|
|
from PIL import Image
|
|
|
|
suffixes = {".png", ".jpg", ".jpeg"}
|
|
sample = sorted(p for p in objects.rglob("*") if p.suffix.lower() in suffixes)[:count]
|
|
if len(sample) < count:
|
|
raise SystemExit(f"{objects}: only {len(sample)} pictures to time the encoder on")
|
|
encoder = ClipEncoder(clip_model)
|
|
began = time.perf_counter()
|
|
rows = 0
|
|
for start in range(0, len(sample), 8):
|
|
batch = [Image.open(path).convert("RGB") for path in sample[start : start + 8]]
|
|
rows += len(encoder.encode_images(batch))
|
|
seconds = time.perf_counter() - began
|
|
if rows != len(sample):
|
|
raise SystemExit(f"the encoder returned {rows} vectors for {len(sample)} pictures")
|
|
return {"pictures": rows, "seconds": seconds, "per_second": rows / seconds}
|
|
|
|
|
|
def _schema(home: Path) -> int:
|
|
import sqlite3
|
|
|
|
with sqlite3.connect(home / "data" / "ely_eye.sqlite") as connection:
|
|
return int(connection.execute("PRAGMA user_version").fetchone()[0])
|
|
|
|
|
|
def rank_all(service: Any, ids: dict[str, str], probes: list[Any], top_k: int) -> tuple[list[list[str]], list[int], list[float]]:
|
|
"""Ask the product as each probe's own holder and collapse atoms to sources by first hit."""
|
|
from ely_eye.schemas import SearchFilters
|
|
|
|
rankings: list[list[str]] = []
|
|
distinct: list[int] = []
|
|
timings: list[float] = []
|
|
for index, query in enumerate(probes):
|
|
began = time.perf_counter()
|
|
hits = service.search(query.text, top_k=top_k, filters=SearchFilters(as_holder=query.holder)).hits
|
|
timings.append(time.perf_counter() - began)
|
|
sources: list[str] = []
|
|
for hit in hits:
|
|
source = ids[hit.atom.source_id]
|
|
if source not in sources:
|
|
sources.append(source)
|
|
distinct.append(len(sources))
|
|
rankings.append(sources)
|
|
if (index + 1) % 100 == 0:
|
|
print(f" {index + 1}/{len(probes)}", flush=True)
|
|
return rankings, distinct, timings
|
|
|
|
|
|
def _groups(
|
|
answerable: list[Any], scores: list[Any], rankings: list[list[str]], times: dict[str, str]
|
|
) -> dict[str, Any]:
|
|
"""recall@10 by modality and by the question type the set recorded, plus who came back first."""
|
|
by_modality: dict[str, list[float]] = {}
|
|
by_category: dict[str, list[float]] = {}
|
|
newest: dict[str, list[bool]] = {}
|
|
every_newest: list[bool] = []
|
|
for query, score, ranking in zip(answerable, scores, rankings, strict=True):
|
|
by_modality.setdefault(query.modality, []).append(score.recall_at_10)
|
|
by_category.setdefault(query.category or "unlabelled", []).append(score.recall_at_10)
|
|
flag = newest_gold_first(ranking, set(query.gold), times)
|
|
if flag is not None:
|
|
newest.setdefault(query.category or "unlabelled", []).append(flag)
|
|
every_newest.append(flag)
|
|
return {
|
|
"by_modality": {name: sum(rows) / len(rows) for name, rows in sorted(by_modality.items())},
|
|
"modality_counts": {name: len(rows) for name, rows in sorted(by_modality.items())},
|
|
"by_category": {name: sum(rows) / len(rows) for name, rows in sorted(by_category.items())},
|
|
"category_counts": {name: len(rows) for name, rows in sorted(by_category.items())},
|
|
"newest_gold_first": sum(every_newest) / len(every_newest) if every_newest else None,
|
|
"newest_gold_first_by_category": {name: sum(rows) / len(rows) for name, rows in sorted(newest.items())},
|
|
"newest_gold_first_counts": {name: len(rows) for name, rows in sorted(newest.items())},
|
|
}
|
|
|
|
|
|
def measure(home: Path, scratch: Path, objects: Path, clip_index: Path, probes: int, models: dict[str, Path]) -> dict[str, Any]:
|
|
"""Score every configuration on the same probes, one warm service per configuration."""
|
|
from ely_eye.config import Settings
|
|
from ely_eye.db import Database
|
|
from ely_eye.retrieval import RetrievalService
|
|
|
|
shipped = Settings().retrieval_top_k
|
|
if shipped != SHIPPED_TOP_K:
|
|
raise SystemExit(
|
|
f"the product asks for {shipped} atoms per search but this harness pins "
|
|
f"SHIPPED_TOP_K={SHIPPED_TOP_K}; update the rows before trusting the 'as shipped' label"
|
|
)
|
|
|
|
every = load_queries()
|
|
answerable = [query for query in every if query.gold]
|
|
if not answerable:
|
|
raise SystemExit("the frozen suite carries no answerable query; nothing to score")
|
|
timed = {query.query_id for query in scale_probes(every, probes) if query.gold}
|
|
if not timed:
|
|
raise SystemExit(
|
|
f"--probes {probes} picked no query to time: the sample strides the suite by modality, "
|
|
"so a small count can skip one and the latency row would be a single modality"
|
|
)
|
|
print(f"{len(answerable)} answerable queries scored, {len(timed)} of them timed", flush=True)
|
|
times = source_times()
|
|
|
|
vectors = _dense_vectors(home / "data" / "ely_eye.sqlite")
|
|
source_schema = _schema(home)
|
|
if vectors == 0 and any(config["dense"] for config in CONFIGS):
|
|
raise SystemExit(
|
|
f"{home}: the store carries no dense vectors, so a dense arm here would rank an empty "
|
|
"index and the 24-minute build would hide outside the measured latency. Build them with "
|
|
"python -m bench.dense_scale, or point --home at a store that already has them."
|
|
)
|
|
|
|
arms: list[dict[str, Any]] = []
|
|
picture: dict[str, Any] | None = None
|
|
rate: dict[str, Any] | None = None
|
|
copy_home = home
|
|
for config in CONFIGS:
|
|
print(f" {config['label']}: dense={config['dense']} visual={config['visual']}", flush=True)
|
|
copy_home = prepare(
|
|
home,
|
|
scratch,
|
|
objects,
|
|
clip_index,
|
|
dense=config["dense"],
|
|
visual=config["visual"],
|
|
clip_model=models["clip"],
|
|
embedding_model=models["embedding"],
|
|
)
|
|
if config["visual"]:
|
|
picture = picture_facts(clip_index)
|
|
if rate is None:
|
|
rate = encode_rate(models["clip"], objects, ENCODE_SAMPLE)
|
|
settings = Settings(
|
|
home=copy_home,
|
|
dense_embeddings_enabled=config["dense"],
|
|
embedding_model_dir=models["embedding"] if config["dense"] else None,
|
|
clip_model_dir=models["clip"] if config["visual"] else None,
|
|
)
|
|
database = Database(settings)
|
|
service = RetrievalService(settings=settings, database=database)
|
|
ids = _source_ids(database)
|
|
rankings, distinct, timings = rank_all(service, ids, answerable, int(config["top_k"]))
|
|
scores = [
|
|
score_query(query.query_id, ranking, set(query.gold), query.holder)
|
|
for query, ranking in zip(answerable, rankings, strict=True)
|
|
]
|
|
summary = aggregate(scores)
|
|
timing = latency([t for query, t in zip(answerable, timings, strict=True) if query.query_id in timed])
|
|
arms.append(
|
|
{
|
|
"label": config["label"],
|
|
"top_k": config["top_k"],
|
|
"dense": config["dense"],
|
|
"visual": config["visual"],
|
|
"recall_at_10": summary["recall_at_10"],
|
|
"mrr": summary["mrr"],
|
|
"hit_at_10": summary["hit_at_10"],
|
|
"wilson_lower": summary["hit_at_10_wilson_lower"],
|
|
"p50_seconds": timing["p50"],
|
|
"p95_seconds": timing["p95"],
|
|
"median_distinct_sources": statistics.median(distinct),
|
|
**_groups(answerable, scores, rankings, times),
|
|
}
|
|
)
|
|
print(
|
|
f" recall@10 {summary['recall_at_10']:.3f}, mrr {summary['mrr']:.3f}, "
|
|
f"p50 {timing['p50'] * 1000:.1f}ms",
|
|
flush=True,
|
|
)
|
|
# Read the version after the arms: the copy arrives at the source's schema and the product migrates it on first open, so the pre-run version is not what was measured.
|
|
return {
|
|
"commit": git_commit(),
|
|
"schema_version": _schema(copy_home),
|
|
"source_schema_version": source_schema,
|
|
"store": str(scratch / "data" / "ely_eye.sqlite"),
|
|
"source_store": str(home),
|
|
"dense_vectors": vectors,
|
|
"picture_index": picture,
|
|
"encode_rate": rate,
|
|
"queries": len(answerable),
|
|
"timed_queries": len(timed),
|
|
"arms": arms,
|
|
"machine": machine_facts(),
|
|
}
|
|
|
|
|
|
def _source_ids(database: Any) -> dict[str, str]:
|
|
"""Store source id to benchmark source id, the same mapping every other arm uses."""
|
|
from bench.systems import _corpus_ids
|
|
|
|
return _corpus_ids(database)
|
|
|
|
|
|
def main() -> None:
|
|
parser = argparse.ArgumentParser(description=__doc__)
|
|
parser.add_argument("--home", type=Path, default=DEFAULT_HOME)
|
|
parser.add_argument("--scratch", type=Path, default=DEFAULT_SCRATCH)
|
|
parser.add_argument("--objects", type=Path, default=DEFAULT_OBJECTS)
|
|
parser.add_argument("--clip-index", type=Path, default=DEFAULT_CLIP_INDEX)
|
|
parser.add_argument("--clip-model", type=Path, default=DEFAULT_CLIP_MODEL)
|
|
parser.add_argument("--embedding-model", type=Path, default=DEFAULT_EMBEDDING_MODEL)
|
|
parser.add_argument("--probes", type=int, default=300)
|
|
parser.add_argument("--out", type=Path, default=Path("bench/reports/LEGS.md"))
|
|
parser.add_argument("--json-out", type=Path, default=Path("bench/reports/leg_check.json"))
|
|
args = parser.parse_args()
|
|
|
|
report = measure(
|
|
args.home,
|
|
args.scratch,
|
|
args.objects,
|
|
args.clip_index,
|
|
args.probes,
|
|
{"clip": args.clip_model, "embedding": args.embedding_model},
|
|
)
|
|
args.json_out.parent.mkdir(parents=True, exist_ok=True)
|
|
args.json_out.write_text(json.dumps(report, indent=2, ensure_ascii=False), encoding="utf-8")
|
|
write_markdown(report, args.out)
|
|
print(f"wrote {args.out}")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
if sys.version_info < (3, 11):
|
|
raise SystemExit("bench needs Python 3.11+")
|
|
main()
|