"""Lightweight placeholder for deep cognitive diagnosis inference.
中文：深度认知诊断（DeepIRT）占位推理逻辑，与路由解耦，生成掌握度向量。
"""
from __future__ import annotations
from typing import Dict, List
from ...pydantic_models import Interaction, MasteryResponse
class _MockDeepIRTModel:
"""代替真实 DeepIRT 的简化版，仅用于演示接口集成。"""
def predict(self, input_data: tuple[List[int], List[int], List[int]]) -> List[float]:
_, _, correctness = input_data
if not correctness:
return [0.5, 0.5, 0.5]
accuracy = sum(correctness) / len(correctness)
# 生成 3 维“知识掌握”向量，略微扰动以模拟不同知识点
return [
max(0.05, min(0.95, accuracy * 0.9 + 0.1)),
max(0.05, min(0.95, accuracy * 0.8 + 0.15)),
max(0.05, min(0.95, accuracy * 1.0)),
]
deepirt_model = _MockDeepIRTModel()
def run_inference_cdm(interactions: List[Interaction]) -> MasteryResponse:
"""输入学生交互序列，输出知识点掌握情况概率。"""
student_ids = [it.student_id for it in interactions]
item_ids = [it.item_id for it in interactions]
correctness = [it.correct for it in interactions]
input_data = (student_ids, item_ids, correctness)
mastery_probs = deepirt_model.predict(input_data)
mastery_dict: Dict[str, float] = {
f"K{idx+1}": float(prob) for idx, prob in enumerate(mastery_probs)
}
return MasteryResponse(mastery=mastery_dict, raw_vector=list(mastery_probs))
