distant supervision的关系抽取代码，带注释

以下是一个简单的关系抽取代码，使用的是distant supervision方法。这个方法是使用预先定义的知识库中的实体和关系标签，将它们对应到包含这些实体的句子中，然后从中提取特征并训练模型进行关系抽取。这种方法可以利用大规模的知识库和文本语料库，但也存在标注错误和知识库不完备等问题。

import os
import re
import numpy as np
import pandas as pd
import torch
import torch.nn as nn
from transformers import AutoTokenizer, AutoModel

# 设置随机种子，以便复现结果
seed = 1234
np.random.seed(seed)
torch.manual_seed(seed)
torch.cuda.manual_seed_all(seed)

# 加载预训练的tokenizer和model
tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
model = AutoModel.from_pretrained('bert-base-uncased')

# 加载数据
data_dir = 'data'
train_df = pd.read_csv(os.path.join(data_dir, 'train.csv'))
test_df = pd.read_csv(os.path.join(data_dir, 'test.csv'))

# 定义正则表达式，以从句子中提取实体和关系标签
entity_regex = re.compile(r'<e>(.*?)<\/e>')
relation_regex = re.compile(r'<r>(.*?)<\/r>')

# 将句子中的实体替换为特殊标记
def replace_entities(sentence, entities):
    for entity in entities:
        sentence = sentence.replace(entity, '<e>' + entity + '</e>')
    return sentence

# 将句子中的关系替换为特殊标记
def replace_relation(sentence, relation):
    return sentence.replace(relation, '<r>' + relation + '</r>')

# 将句子转换为tokens，并使用特殊的CLS和SEP token
def tokenize(sentence):
    tokens = tokenizer.tokenize(sentence)
    tokens = ['[CLS]'] + tokens + ['[SEP]']
    return tokens

# 将tokens转换为IDs
def convert_to_ids(tokens):
    input_ids = tokenizer.convert_tokens_to_ids(tokens)
    return input_ids

# 将tokens对齐到指定长度，并在不足长度时用0填充
def pad_sequence(sequence, max_len):
    if len(sequence) < max_len:
        sequence += [0] * (max_len - len(sequence))
    else:
        sequence = sequence[:max_len]
    return sequence

# 为数据集创建一个Dataset类，用于在训练时读取数据
class RelationExtractionDataset(torch.utils.data.Dataset):
    def __init__(self, df):
        self.df = df
        
    def __len__(self):
        return len(self.df)
    
    def __getitem__(self, index):
        row = self.df.iloc[index]
        sentence = row['sentence']
        entities = row['entities'].split(',')
        relation = row['relation']
        
        # 将实体替换为特殊标记
        sentence = replace_entities(sentence, entities)
        
        # 将关系替换为特殊标记
        sentence = replace_relation(sentence, relation)
        
        # 将句子转换为tokens
        tokens = tokenize(sentence)
        
        # 将tokens转换