feat(question_gen): sample_anchor — 按题型层级采样锚节点
含 6 种层级分支:L3 单帧、L2 多帧、Temporal Perception 特例、 L1 全量/采样 L2、L1-L2 混合。时间排序 + used_node_ids 排除。 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -7,6 +7,12 @@
|
||||
from __future__ import annotations
|
||||
|
||||
from dataclasses import dataclass
|
||||
from typing import TYPE_CHECKING
|
||||
|
||||
if TYPE_CHECKING:
|
||||
import random
|
||||
|
||||
from app.tree.index import L1Node, L2Node, L3Node, TreeIndex
|
||||
|
||||
|
||||
@dataclass(frozen=True)
|
||||
@@ -66,3 +72,396 @@ TASK_TYPE_LEVEL_MAP: dict[str, TaskTypeSpec] = {
|
||||
"Information Synopsis": TaskTypeSpec("L1", True, "per-L2", ("scene_summary",)),
|
||||
"Object Reasoning": TaskTypeSpec("L1-L2", True, "per-L2", ("event_description",)),
|
||||
}
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 内部辅助函数
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def _serialize_l3_card(l3: L3Node, context_fields: tuple[str, ...]) -> str:
|
||||
"""将 L3 节点 card 按 context_fields 序列化为可读文本。
|
||||
|
||||
参数:
|
||||
l3: L3 节点。
|
||||
context_fields: 需提取的字段名元组。
|
||||
|
||||
返回:
|
||||
多行 "field: value" 格式的文本。
|
||||
"""
|
||||
parts: list[str] = []
|
||||
for fld in context_fields:
|
||||
val = getattr(l3.card, fld, None)
|
||||
if val is not None:
|
||||
parts.append(f"{fld}: {val}")
|
||||
return "\n".join(parts)
|
||||
|
||||
|
||||
def _l2_time_range_str(l2: L2Node) -> str:
|
||||
"""将 L2 的 time_range 格式化为可读字符串。
|
||||
|
||||
参数:
|
||||
l2: L2 节点。
|
||||
|
||||
返回:
|
||||
"time_range: (start, end)" 格式,或 "time_range: unknown"。
|
||||
"""
|
||||
if l2.time_range is not None:
|
||||
return f"time_range: ({l2.time_range[0]:.2f}, {l2.time_range[1]:.2f})"
|
||||
return "time_range: unknown"
|
||||
|
||||
|
||||
def _representative_frame(l2: L2Node) -> str | None:
|
||||
"""取 L2 的代表帧路径——第一个有 frame_path 的 L3 子节点。
|
||||
|
||||
参数:
|
||||
l2: L2 节点。
|
||||
|
||||
返回:
|
||||
帧路径字符串,或 None(无可用帧时)。
|
||||
"""
|
||||
for l3 in l2.children:
|
||||
if l3.frame_path:
|
||||
return l3.frame_path
|
||||
return None
|
||||
|
||||
|
||||
def _sort_l2_by_time(l2_nodes: list[L2Node]) -> list[L2Node]:
|
||||
"""按 time_range 升序排列 L2 节点(None 排末尾)。
|
||||
|
||||
参数:
|
||||
l2_nodes: 待排序的 L2 节点列表。
|
||||
|
||||
返回:
|
||||
排序后的新列表(不修改原列表)。
|
||||
"""
|
||||
return sorted(
|
||||
l2_nodes,
|
||||
key=lambda n: n.time_range[0] if n.time_range is not None else float("inf"),
|
||||
)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 各层级采样策略
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def _sample_l3(
|
||||
tree: TreeIndex,
|
||||
task_type: str,
|
||||
spec: TaskTypeSpec,
|
||||
used_node_ids: set[str],
|
||||
rng: random.Random,
|
||||
) -> AnchorContext:
|
||||
"""L3 层级锚节点采样。
|
||||
|
||||
收集全部 L3 节点,排除已用节点,随机选取一个。
|
||||
|
||||
参数:
|
||||
tree: 三层树索引。
|
||||
task_type: 题型名称。
|
||||
spec: 题型规格。
|
||||
used_node_ids: 已用节点 ID 集合。
|
||||
rng: 随机数生成器。
|
||||
|
||||
返回:
|
||||
AnchorContext 实例。
|
||||
|
||||
异常:
|
||||
ValueError: 候选 L3 节点不足。
|
||||
"""
|
||||
# Phase 1: 收集所有 L3 候选
|
||||
candidates: list[tuple[L3Node, L2Node]] = []
|
||||
for root in tree.roots:
|
||||
for l2 in root.children:
|
||||
for l3 in l2.children:
|
||||
if l3.id not in used_node_ids:
|
||||
candidates.append((l3, l2))
|
||||
|
||||
if not candidates:
|
||||
raise ValueError(f"锚节点不足: {task_type} 无可用 L3 节点")
|
||||
|
||||
# Phase 2: 随机选取
|
||||
chosen_l3, parent_l2 = rng.choice(candidates)
|
||||
|
||||
# Phase 3: 构造上下文
|
||||
card_text = _serialize_l3_card(chosen_l3, spec.context_fields)
|
||||
frame_paths = [chosen_l3.frame_path] if chosen_l3.frame_path else []
|
||||
subtitle = chosen_l3.subtitle or ""
|
||||
|
||||
# Phase 4: 干扰项——同 L2 下其他 L3 的 frame_summary
|
||||
distractor_texts = [l3.card.frame_summary for l3 in parent_l2.children if l3.id != chosen_l3.id]
|
||||
|
||||
return AnchorContext(
|
||||
node_id=chosen_l3.id,
|
||||
card_text=card_text,
|
||||
frame_paths=frame_paths,
|
||||
subtitle=subtitle,
|
||||
distractor_texts=distractor_texts,
|
||||
)
|
||||
|
||||
|
||||
def _sample_l2(
|
||||
tree: TreeIndex,
|
||||
task_type: str,
|
||||
spec: TaskTypeSpec,
|
||||
used_node_ids: set[str],
|
||||
rng: random.Random,
|
||||
) -> AnchorContext:
|
||||
"""L2 层级锚节点采样(含 Temporal Perception 特殊处理)。
|
||||
|
||||
普通 L2 题型:随机选 1 个 L2,取 2-3 个子 L3 帧。
|
||||
Temporal Perception:0-1 帧,card_text 必含 time_range。
|
||||
|
||||
参数:
|
||||
tree: 三层树索引。
|
||||
task_type: 题型名称。
|
||||
spec: 题型规格。
|
||||
used_node_ids: 已用节点 ID 集合。
|
||||
rng: 随机数生成器。
|
||||
|
||||
返回:
|
||||
AnchorContext 实例。
|
||||
|
||||
异常:
|
||||
ValueError: 候选 L2 节点不足。
|
||||
"""
|
||||
# Phase 1: 收集所有 L2 候选
|
||||
all_l2: list[tuple[L2Node, L1Node]] = []
|
||||
for root in tree.roots:
|
||||
for l2 in root.children:
|
||||
if l2.id not in used_node_ids:
|
||||
all_l2.append((l2, root))
|
||||
|
||||
if not all_l2:
|
||||
raise ValueError(f"锚节点不足: {task_type} 无可用 L2 节点")
|
||||
|
||||
# Phase 2: 随机选取
|
||||
chosen_l2, parent_l1 = rng.choice(all_l2)
|
||||
|
||||
is_temporal_perception = task_type == "Temporal Perception"
|
||||
|
||||
# Phase 3: 帧路径
|
||||
if is_temporal_perception:
|
||||
# 0-1 帧:有子节点则取 1 帧,否则 0 帧
|
||||
frame_paths: list[str] = []
|
||||
if chosen_l2.children:
|
||||
first_frame = chosen_l2.children[0].frame_path
|
||||
if first_frame:
|
||||
frame_paths = [first_frame]
|
||||
else:
|
||||
# 普通 L2:随机采样 2-3 个 L3 帧
|
||||
children_with_frames = [l3 for l3 in chosen_l2.children if l3.frame_path]
|
||||
n_frames = min(rng.randint(2, 3), len(children_with_frames))
|
||||
sampled = rng.sample(children_with_frames, n_frames) if n_frames > 0 else []
|
||||
frame_paths = [l3.frame_path for l3 in sampled if l3.frame_path]
|
||||
|
||||
# Phase 4: card_text
|
||||
card_text = f"event_description: {chosen_l2.card.event_description}"
|
||||
if is_temporal_perception:
|
||||
card_text += f"\n{_l2_time_range_str(chosen_l2)}"
|
||||
|
||||
# Phase 5: 字幕(L2 无自身字幕,取首个子 L3 字幕)
|
||||
subtitle = ""
|
||||
if chosen_l2.children and chosen_l2.children[0].subtitle:
|
||||
subtitle = chosen_l2.children[0].subtitle
|
||||
|
||||
# Phase 6: 干扰项——同 L1 下其他 L2 的 event_description
|
||||
distractor_texts = [
|
||||
l2.card.event_description for l2 in parent_l1.children if l2.id != chosen_l2.id
|
||||
]
|
||||
|
||||
return AnchorContext(
|
||||
node_id=chosen_l2.id,
|
||||
card_text=card_text,
|
||||
frame_paths=frame_paths,
|
||||
subtitle=subtitle,
|
||||
distractor_texts=distractor_texts,
|
||||
)
|
||||
|
||||
|
||||
def _sample_l1(
|
||||
tree: TreeIndex,
|
||||
task_type: str,
|
||||
spec: TaskTypeSpec,
|
||||
used_node_ids: set[str],
|
||||
rng: random.Random,
|
||||
) -> AnchorContext:
|
||||
"""L1 层级锚节点采样(Temporal Reasoning / Information Synopsis)。
|
||||
|
||||
Information Synopsis:使用目标 L1 下全部 L2 子节点。
|
||||
Temporal Reasoning:使用 >=3 个 L2 子节点(不足 3 个则全部使用)。
|
||||
L2 按 time_range 升序排列,每个 L2 取一帧代表。
|
||||
|
||||
参数:
|
||||
tree: 三层树索引。
|
||||
task_type: 题型名称。
|
||||
spec: 题型规格。
|
||||
used_node_ids: 已用节点 ID 集合。
|
||||
rng: 随机数生成器。
|
||||
|
||||
返回:
|
||||
AnchorContext 实例。
|
||||
|
||||
异常:
|
||||
ValueError: 候选 L1 节点不足。
|
||||
"""
|
||||
# Phase 1: 收集可用 L1
|
||||
candidates = [r for r in tree.roots if r.id not in used_node_ids]
|
||||
if not candidates:
|
||||
raise ValueError(f"锚节点不足: {task_type} 无可用 L1 节点")
|
||||
|
||||
# Phase 2: 随机选取
|
||||
chosen_l1 = rng.choice(candidates)
|
||||
|
||||
# Phase 3: 选定 L2 子集
|
||||
if task_type == "Information Synopsis":
|
||||
# 必须使用全部 L2
|
||||
selected_l2 = list(chosen_l1.children)
|
||||
else:
|
||||
# Temporal Reasoning:>=3 个 L2(不足则全部)
|
||||
if len(chosen_l1.children) <= 3:
|
||||
selected_l2 = list(chosen_l1.children)
|
||||
else:
|
||||
selected_l2 = rng.sample(chosen_l1.children, rng.randint(3, len(chosen_l1.children)))
|
||||
|
||||
# Phase 4: 按 time_range 升序排列
|
||||
selected_l2 = _sort_l2_by_time(selected_l2)
|
||||
|
||||
# Phase 5: card_text(场景摘要)
|
||||
card_text = f"scene_summary: {chosen_l1.card.scene_summary}"
|
||||
|
||||
# Phase 6: 帧路径——每个 L2 取一帧代表
|
||||
frame_paths: list[str] = []
|
||||
for l2 in selected_l2:
|
||||
rep = _representative_frame(l2)
|
||||
if rep:
|
||||
frame_paths.append(rep)
|
||||
|
||||
# Phase 7: 字幕(L1 无字幕)
|
||||
subtitle = ""
|
||||
|
||||
# Phase 8: 干扰项——其他 L1 的 scene_summary
|
||||
distractor_texts = [r.card.scene_summary for r in tree.roots if r.id != chosen_l1.id]
|
||||
|
||||
return AnchorContext(
|
||||
node_id=chosen_l1.id,
|
||||
card_text=card_text,
|
||||
frame_paths=frame_paths,
|
||||
subtitle=subtitle,
|
||||
distractor_texts=distractor_texts,
|
||||
)
|
||||
|
||||
|
||||
def _sample_l1_l2(
|
||||
tree: TreeIndex,
|
||||
task_type: str,
|
||||
spec: TaskTypeSpec,
|
||||
used_node_ids: set[str],
|
||||
rng: random.Random,
|
||||
) -> AnchorContext:
|
||||
"""L1-L2 跨层级锚节点采样(Object Reasoning)。
|
||||
|
||||
从全部 L2 中随机选 2-3 个,按 time_range 排序,
|
||||
card_text 为各 L2 的 event_description 拼接。
|
||||
|
||||
参数:
|
||||
tree: 三层树索引。
|
||||
task_type: 题型名称。
|
||||
spec: 题型规格。
|
||||
used_node_ids: 已用节点 ID 集合。
|
||||
rng: 随机数生成器。
|
||||
|
||||
返回:
|
||||
AnchorContext 实例。
|
||||
|
||||
异常:
|
||||
ValueError: 候选 L2 节点不足。
|
||||
"""
|
||||
# Phase 1: 收集全部 L2
|
||||
all_l2: list[L2Node] = []
|
||||
for root in tree.roots:
|
||||
for l2 in root.children:
|
||||
if l2.id not in used_node_ids:
|
||||
all_l2.append(l2)
|
||||
|
||||
if not all_l2:
|
||||
raise ValueError(f"锚节点不足: {task_type} 无可用 L2 节点")
|
||||
|
||||
# Phase 2: 随机选 2-3 个
|
||||
n_pick = min(rng.randint(2, 3), len(all_l2))
|
||||
selected = rng.sample(all_l2, n_pick)
|
||||
|
||||
# Phase 3: 按 time_range 升序排列
|
||||
selected = _sort_l2_by_time(selected)
|
||||
|
||||
# Phase 4: card_text = 各 L2 event_description 拼接
|
||||
card_text = "\n".join(f"event_description: {l2.card.event_description}" for l2 in selected)
|
||||
|
||||
# Phase 5: 帧路径——每个 L2 取一帧代表
|
||||
frame_paths: list[str] = []
|
||||
for l2 in selected:
|
||||
rep = _representative_frame(l2)
|
||||
if rep:
|
||||
frame_paths.append(rep)
|
||||
|
||||
# Phase 6: 字幕
|
||||
subtitle = ""
|
||||
|
||||
# Phase 7: 干扰项——未被选中的 L2 的 event_description
|
||||
selected_ids = {l2.id for l2 in selected}
|
||||
distractor_texts = [l2.card.event_description for l2 in all_l2 if l2.id not in selected_ids]
|
||||
|
||||
# 使用第一个被选中节点的 ID 作为锚节点 ID
|
||||
anchor_id = selected[0].id
|
||||
|
||||
return AnchorContext(
|
||||
node_id=anchor_id,
|
||||
card_text=card_text,
|
||||
frame_paths=frame_paths,
|
||||
subtitle=subtitle,
|
||||
distractor_texts=distractor_texts,
|
||||
)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 公开接口
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def sample_anchor(
|
||||
tree: TreeIndex,
|
||||
task_type: str,
|
||||
used_node_ids: set[str],
|
||||
rng: random.Random,
|
||||
) -> AnchorContext:
|
||||
"""根据题型从视频树中采样锚节点及上下文素材。
|
||||
|
||||
依据 TASK_TYPE_LEVEL_MAP 中的层级规格,分发到对应的层级采样策略。
|
||||
每种层级有不同的帧选取、card 序列化和干扰项收集逻辑。
|
||||
|
||||
参数:
|
||||
tree: 三层树索引。
|
||||
task_type: 12 种 Video-MME 题型之一。
|
||||
used_node_ids: 本轮已用节点 ID 集合(避免重复采样)。
|
||||
rng: 可控随机数生成器(保证可复现)。
|
||||
|
||||
返回:
|
||||
AnchorContext 实例,包含锚节点 ID、card 文本、帧路径、字幕和干扰项。
|
||||
|
||||
异常:
|
||||
KeyError: task_type 不在 TASK_TYPE_LEVEL_MAP 中。
|
||||
ValueError: 候选节点不足(全部被 used_node_ids 排除)。
|
||||
"""
|
||||
spec = TASK_TYPE_LEVEL_MAP[task_type]
|
||||
|
||||
if spec.level == "L3":
|
||||
return _sample_l3(tree, task_type, spec, used_node_ids, rng)
|
||||
elif spec.level == "L2":
|
||||
return _sample_l2(tree, task_type, spec, used_node_ids, rng)
|
||||
elif spec.level == "L1":
|
||||
return _sample_l1(tree, task_type, spec, used_node_ids, rng)
|
||||
elif spec.level == "L1-L2":
|
||||
return _sample_l1_l2(tree, task_type, spec, used_node_ids, rng)
|
||||
else:
|
||||
raise ValueError(f"未知层级: {spec.level}")
|
||||
|
||||
@@ -1,10 +1,20 @@
|
||||
"""synthesizer 模块单元测试 — AnchorContext + 题型映射常量。"""
|
||||
"""synthesizer 模块单元测试 — AnchorContext + 题型映射常量 + sample_anchor。"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import dataclasses
|
||||
import random
|
||||
from pathlib import Path
|
||||
|
||||
from app.question_gen.synthesizer import TASK_TYPE_LEVEL_MAP, AnchorContext, TaskTypeSpec
|
||||
import pytest
|
||||
|
||||
from app.question_gen.synthesizer import (
|
||||
TASK_TYPE_LEVEL_MAP,
|
||||
AnchorContext,
|
||||
TaskTypeSpec,
|
||||
sample_anchor,
|
||||
)
|
||||
from app.tree.index import TreeIndex
|
||||
|
||||
ALL_12_TYPES = [
|
||||
"Object Recognition",
|
||||
@@ -132,3 +142,89 @@ class TestTaskTypeSpec:
|
||||
assert isinstance(spec.context_fields, tuple), (
|
||||
f"{task_type} 的 context_fields 不是 tuple"
|
||||
)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# sample_anchor 测试
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def _load_test_tree() -> tuple[TreeIndex, str]:
|
||||
"""加载真实测试树(store/videos/ 下第一棵)。"""
|
||||
videos_dir = Path("store/videos")
|
||||
first_vid = sorted(videos_dir.iterdir())[0]
|
||||
tree = TreeIndex.load_json(str(first_vid / "tree.json"))
|
||||
return tree, first_vid.name
|
||||
|
||||
|
||||
class TestSampleAnchor:
|
||||
"""sample_anchor 锚节点采样测试(基于真实树数据)。"""
|
||||
|
||||
def test_l3_type_returns_single_frame(self) -> None:
|
||||
"""L3 题型(Object Recognition)应返回恰好 1 帧。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
ctx = sample_anchor(tree, "Object Recognition", set(), random.Random(42))
|
||||
assert len(ctx.frame_paths) == 1
|
||||
assert ctx.node_id.startswith("L") or "_L3_" in ctx.node_id
|
||||
assert len(ctx.distractor_texts) > 0
|
||||
|
||||
def test_l2_type_returns_multiple_frames(self) -> None:
|
||||
"""L2 题型(Action Reasoning)应返回 2-3 帧。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
ctx = sample_anchor(tree, "Action Reasoning", set(), random.Random(42))
|
||||
assert 2 <= len(ctx.frame_paths) <= 3
|
||||
|
||||
def test_temporal_perception_zero_or_one_frame(self) -> None:
|
||||
"""Temporal Perception 特殊处理:0-1 帧,且 card_text 包含 time_range。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
ctx = sample_anchor(tree, "Temporal Perception", set(), random.Random(42))
|
||||
assert len(ctx.frame_paths) <= 1
|
||||
assert "time_range" in ctx.card_text.lower() or "time" in ctx.card_text.lower()
|
||||
|
||||
def test_information_synopsis_uses_all_l2(self) -> None:
|
||||
"""Information Synopsis 必须使用目标 L1 下所有 L2 子节点。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
ctx = sample_anchor(tree, "Information Synopsis", set(), random.Random(42))
|
||||
# 至少应有帧(每个 L2 取一帧代表)
|
||||
total_l2 = sum(len(r.children) for r in tree.roots)
|
||||
assert len(ctx.frame_paths) >= min(total_l2, 1)
|
||||
|
||||
def test_l1_type_l2_nodes_in_time_order(self) -> None:
|
||||
"""L1 题型(Temporal Reasoning)的 card_text 应有实质内容。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
ctx = sample_anchor(tree, "Temporal Reasoning", set(), random.Random(42))
|
||||
assert len(ctx.frame_paths) >= 1
|
||||
assert len(ctx.card_text) > 20
|
||||
|
||||
def test_used_node_ids_excluded(self) -> None:
|
||||
"""used_node_ids 中的节点不应被再次选中。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
rng = random.Random(42)
|
||||
ctx1 = sample_anchor(tree, "Object Recognition", set(), rng)
|
||||
ctx2 = sample_anchor(tree, "Object Recognition", {ctx1.node_id}, random.Random(43))
|
||||
assert ctx2.node_id != ctx1.node_id
|
||||
|
||||
def test_insufficient_nodes_raises(self) -> None:
|
||||
"""所有候选节点均被排除时应抛出 ValueError。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
all_l3_ids: set[str] = set()
|
||||
for root in tree.roots:
|
||||
for l2 in root.children:
|
||||
for l3 in l2.children:
|
||||
all_l3_ids.add(l3.id)
|
||||
with pytest.raises(ValueError, match="锚节点不足"):
|
||||
sample_anchor(tree, "Object Recognition", all_l3_ids, random.Random(42))
|
||||
|
||||
def test_object_reasoning_l1_l2_type(self) -> None:
|
||||
"""Object Reasoning (L1-L2) 应选 2-3 个 L2 并按时间排序。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
ctx = sample_anchor(tree, "Object Reasoning", set(), random.Random(42))
|
||||
assert 1 <= len(ctx.frame_paths) <= 3
|
||||
assert len(ctx.card_text) > 10
|
||||
|
||||
def test_spatial_reasoning_context_fields(self) -> None:
|
||||
"""Spatial Reasoning 的 card_text 应包含 spatial_layout 内容。"""
|
||||
tree, _vid = _load_test_tree()
|
||||
ctx = sample_anchor(tree, "Spatial Reasoning", set(), random.Random(42))
|
||||
# context_fields 包含 spatial_layout,card_text 应含有该字段内容
|
||||
assert len(ctx.card_text) > 20
|
||||
|
||||
Reference in New Issue
Block a user