解决pandas报错single positional indexer is out-of-bounds 跳过df3缺失词汇计算得分

报错原因 #

当句子中出现的词汇不在df3的Word列范围内时，df3[df3.Word == word]会返回空DataFrame，此时调用.iloc[0]取第一行数据的操作就会触发IndexError: single positional indexer is out-of-bounds报错。

解决方案 #

方案1：最小改动原有代码 #

只需在查询属性前先判断匹配结果是否为空，为空直接跳过当前词即可：

pos_score = 0
neg_score = 0
for i in range(len(filtered_sentence)):
    for word in filtered_sentence[i]:
        # 先查询匹配结果
        match_result = df3[df3.Word == word]
        # 无匹配结果直接跳过当前词
        if len(match_result) == 0:
            continue
        # 有匹配结果再走原有逻辑
        if match_result.iloc[0]['Negative'] == 2009:
            neg_score = neg_score + 1
        elif match_result.iloc[0]['Positive'] == 2009:
            pos_score = pos_score + 1
        else:
            break

方案2：性能优化版（推荐） #

循环中反复过滤DataFrame效率极低，提前把df3转为字典映射，单次查询时间复杂度直接降到O(1)，数据量越大优势越明显：

pos_score = 0
neg_score = 0
# 提前构建词-属性映射字典
word_attr_map = df3.set_index('Word').to_dict('index')

for i in range(len(filtered_sentence)):
    for word in filtered_sentence[i]:
        # 词不在映射中直接跳过
        if word not in word_attr_map:
            continue
        attr = word_attr_map[word]
        if attr['Negative'] == 2009:
            neg_score += 1
        elif attr['Positive'] == 2009:
            pos_score += 1
        else:
            break

内容的提问来源于stack exchange，提问作者Shreyanshu