使用pgmpy构建贝叶斯信念网络时循环停滞问题求助

我用pgmpy库在Python中构建了贝叶斯信念网络（Bayesian Belief Network），数据集包含5列（Healthy、Growth、Refined、Reasoned、Accepted）共5万行，已划分为1万行训练集和剩余的验证集。网络边定义为：('Healthy', 'Refined')、('Healthy', 'Reasoned')、('Refined', 'Accepted')、('Reasoned', 'Accepted')、('Growth', 'Accepted')。

为评估网络质量，我编写循环从验证集提取Healthy、Growth、Refined、Reasoned节点的证据，预测Accepted节点值并与实际值对比，但循环总是在584次迭代后停滞，无报错且内核仍显示忙碌。以下是采用最大似然估计（Maximum likelihood）方法的简化代码，其他参数计算方法也存在相同问题：

import pandas as pd
from pgmpy.base import DAG 
from pgmpy.models import BayesianNetwork
from pgmpy.sampling import BayesianModelSampling
from pgmpy.factors.discrete import State

#import dataset
df = pd.read_csv("C:\\Users\\puddu\\Desktop\\Tools\\Dummy.BBN\\Dummy_data_set.csv")

#preliminary operation on dataset
df.rename(columns = {'Q1.Healthy':'Healthy', 'Q2.Growth':'Growth',
                              'Q3.Refined':'Refined', 'Q9.Accepted':'Accepted',
                              'Q8.Reasoned':'Reasoned'}, inplace = True)

nodes = ('Healthy', 'Growth', 'Refined', 'Reasoned', 'Accepted')

replies = ['E','D', 'C', 'B', 'A']

edges = [('Healthy', 'Refined'),
          ('Healthy', 'Reasoned'),
          ('Refined', 'Accepted'),
          ('Reasoned', 'Accepted'),
          ('Growth', 'Accepted')]

for nod in nodes:
    df[nod]=df[nod].astype('category')
    df[nod] = df[nod].cat.set_categories(replies, ordered=True) 

#training set definition
df_train = df.head(10000).copy().reset_index(drop= True)

#directed acyclic graph building
dag = DAG()

dag.add_edges_from(ebunch= edges)

#BBN building + estimating MLE parameters
model_mle = BayesianNetwork(dag)

model_mle.fit(df_train)

df_validation = df.iloc[(10000):(11000),].copy().reset_index(drop= True)
inference_mle = BayesianModelSampling(model_mle)
mle_guesses = 0 
for i in range(1000):
    evidence = [State(var= 'Growth', state= df_validation['Growth'][i]),
                State(var= 'Healthy', state= df_validation['Healthy'][i]),
                State(var= 'Reasoned', state= df_validation['Reasoned'][i]),
                State(var= 'Refined', state = df_validation['Refined'][i])]
    mle_prediction = inference_mle.rejection_sample(size= 1,
                     evidence = evidence, show_progress= False)['Accepted'][0]
    result = df_validation['Accepted'][i]
    if mle_prediction == result:
                 mle_guesses+= 1
    print(f"Step {i}") 

问题根源与解决方法 #

核心问题：拒绝采样的低效性 #

停滞的原因是使用了rejection_sample方法。拒绝采样的逻辑是生成大量随机样本，再过滤掉不符合证据的样本——当某组证据的联合概率极低（甚至在训练集中从未出现）时，采样过程会陷入无限等待，因为很难抽到符合条件的样本。你遇到的584次迭代后停滞，大概率是第585条验证数据的证据组合在训练集中出现频率极低，导致采样无法快速找到匹配样本。

替换方案：使用精确推理 #

改用pgmpy的VariableElimination精确推理类，它直接计算后验概率，无需采样，速度更快且不会卡住：

from pgmpy.inference import VariableElimination

# 替换采样推理为精确推理
inference_mle = VariableElimination(model_mle)
mle_guesses = 0 

for i in range(1000):
    # 用字典构造证据，比State列表更简洁
    evidence = {
        'Growth': df_validation['Growth'][i],
        'Healthy': df_validation['Healthy'][i],
        'Reasoned': df_validation['Reasoned'][i],
        'Refined': df_validation['Refined'][i]
    }
    # 计算Accepted的后验概率分布
    posterior = inference_mle.query(variables=['Accepted'], evidence=evidence)
    # 取概率最高的状态作为预测值
    mle_prediction = posterior.state_names['Accepted'][posterior.values.argmax()]
    
    result = df_validation['Accepted'][i]
    if mle_prediction == result:
        mle_guesses += 1
    print(f"Step {i}")

额外优化建议 #

• 批量处理：若验证集规模大，可按证据组合分组批量计算后验概率，避免逐行循环的开销。
• 过滤罕见组合：提前统计验证集中各证据组合的出现频率，对训练集中未出现的组合，可直接标记为未知或使用默认值，避免推理卡壳。

内容的提问来源于stack exchange，提问作者Edoardo Puddu