Python列表处理：提取各标签类别下的首条标签详情

• 按label字段做类别去重，每个标签类别仅保留列表中出现的第一条记录
• 用集合存储已经处理过的标签值，遍历原始列表时跳过同标签的后续记录
• 时间复杂度为O(n)，仅需一次遍历即可完成处理，性能最优

# 原始标签数据
input_data = [
    {'label': 'Accord_row_loc', 'xmin': 48.0, 'ymin': 833.0, 'xmax': 1652.0, 'ymax': 900.3014907836914, 'likelihood': 5},
    {'label': 'Accord_row_loc', 'xmin': 48.0, 'ymin': 900.30078125, 'xmax': 1652.0, 'ymax': 967.30078125, 'likelihood': 5},
    {'label': 'Accord_row_loc', 'xmin': 48.0, 'ymin': 967.421875, 'xmax': 1652.0, 'ymax': 1035.0, 'likelihood': 5},
    {'label': 'Accord_row_contact_info', 'xmin': 170.0, 'ymin': 1583.1669921875, 'xmax': 1651.0, 'ymax': 1617.0, 'likelihood': 5},
    {'label': 'Accord_row_contact_info', 'xmin': 170.0, 'ymin': 1617.0, 'xmax': 1651.0, 'ymax': 1649.1640625, 'likelihood': 5},
    {'label': 'Accord_row_contact_info', 'xmin': 170.1005859375, 'ymin': 1649.2998046875, 'xmax': 1651.0, 'ymax': 1685.0, 'likelihood': 5},
    {'label': 'Accord_row_individuals', 'xmin': 48.0, 'ymin': 1801.0, 'xmax': 1652.0, 'ymax': 1867.0, 'likelihood': 5}
]

def extract_first_per_label(label_list):
    processed_labels = set()
    result = []
    for item in label_list:
        label_val = item['label']
        if label_val not in processed_labels:
            processed_labels.add(label_val)
            result.append(item)
    return result

# 执行处理
output = extract_first_per_label(input_data)

执行代码后得到的输出与预期完全匹配：

[
    {'label': 'Accord_row_loc', 'xmin': 48.0, 'ymin': 833.0, 'xmax': 1652.0, 'ymax': 900.3014907836914, 'likelihood': 5},
    {'label': 'Accord_row_contact_info', 'xmin': 170.0, 'ymin': 1583.1669921875, 'xmax': 1651.0, 'ymax': 1617.0, 'likelihood': 5},
    {'label': 'Accord_row_individuals', 'xmin': 48.0, 'ymin': 1801.0, 'xmax': 1652.0, 'ymax': 1867.0, 'likelihood': 5}
]

注：代码中避免使用input作为变量名，防止覆盖Python内置的input()函数引发异常。

内容的提问来源于stack exchange，提问作者Zeeshan Akhtar