將包含許多嵌套對象的 JSON 文件保存到列表中

Question

我目前有一個具有以下格式的 JSON 文件。 請記住，這不是整個文件。 整個 JSON 文件由數百個鍵值對列表組成，這些鍵值對在方括號 [][][]... 之間一個接一個。 我試圖將這些單獨的列表中的每一個存儲在一個結構中，例如下面的列表，以便我可以迭代結構並解析每個列表的文件名、標簽等。我最初嘗試使用 json.loads() 導入它，但我有由於對象的嵌套性質而導致的問題。 我將不勝感激有關如何將此文件導入列表或其他適當的 Python 結構的任何想法/見解。

[
   {
      "File_Name": "1.jpg",
      "Analysis": {
         "Labels": [
            {
               "Confidence": 94.77251434326172,
               "Name": "Flora"
            },
            {
               "Confidence": 94.77251434326172,
               "Name": "Grass"
            },
            {
               "Confidence": 94.77251434326172,
               "Name": "Plant"
            },
            {
               "Confidence": 78.49254608154297,
               "Name": "Animal"
            },
            {
               "Confidence": 78.49254608154297,
               "Name": "Cheetah"
            },
            {
               "Confidence": 78.49254608154297,
               "Name": "Mammal"
            },
            {
               "Confidence": 78.49254608154297,
               "Name": "Wildlife"
            },
            {
               "Confidence": 69.79740142822266,
               "Name": "Field"
            },
            {
               "Confidence": 69.79740142822266,
               "Name": "Grassland"
            },
            {
               "Confidence": 69.79740142822266,
               "Name": "Outdoors"
            },
            {
               "Confidence": 67.31356048583984,
               "Name": "Leisure Activities"
            },
            {
               "Confidence": 67.31356048583984,
               "Name": "Walking"
            },
            {
               "Confidence": 57.44683837890625,
               "Name": "Jaguar"
            },
            {
               "Confidence": 57.44683837890625,
               "Name": "Leopard"
            },
            {
               "Confidence": 57.44683837890625,
               "Name": "Panther"
            },
            {
               "Confidence": 55.88261032104492,
               "Name": "Bush"
            },
            {
               "Confidence": 55.88261032104492,
               "Name": "Vegetation"
            },
            {
               "Confidence": 53.4413948059082,
               "Name": "Lawn"
            }
         ],
         "ResponseMetadata": {
            "RetryAttempts": 0,
            "HTTPStatusCode": 200,
            "RequestId": "978e32e4-1da8-11e8-a380-cd680f89684e",
            "HTTPHeaders": {
               "date": "Thu, 01 Mar 2018 23:30:59 GMT",
               "x-amzn-requestid": "978e32e4-1da8-11e8-a380-cd680f89684e",
               "content-length": "947",
               "content-type": "application/x-amz-json-1.1",
               "connection": "keep-alive"
            }
         },
         "OrientationCorrection": "ROTATE_0"
      }
   }
][
   {
      "File_Name": "2.jpg",
      "Analysis": {
         "Labels": [
            {
               "Confidence": 98.57389068603516,
               "Name": "Astronomy"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Galaxy"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Nebula"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Night"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Outdoors"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Outer Space"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Space"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Universe"
            }
         ],
         "ResponseMetadata": {
            "RetryAttempts": 0,
            "HTTPStatusCode": 200,
            "RequestId": "98d2c109-1da8-11e8-a2d9-b91cf22c7f33",
            "HTTPHeaders": {
               "date": "Thu, 01 Mar 2018 23:30:59 GMT",
               "x-amzn-requestid": "98d2c109-1da8-11e8-a2d9-b91cf22c7f33",
               "content-length": "449",
               "content-type": "application/x-amz-json-1.1",
               "connection": "keep-alive"
            }
         },
         "OrientationCorrection": "ROTATE_0"
      }
   },
   {
      "File_Name": "2.jpg",
      "Analysis": {
         "Labels": [
            {
               "Confidence": 98.57389068603516,
               "Name": "Astronomy"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Galaxy"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Nebula"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Night"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Outdoors"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Outer Space"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Space"
            },
            {
               "Confidence": 98.57389068603516,
               "Name": "Universe"
            }
         ],
         "ResponseMetadata": {
            "RetryAttempts": 0,
            "HTTPStatusCode": 200,
            "RequestId": "98d2c109-1da8-11e8-a2d9-b91cf22c7f33",
            "HTTPHeaders": {
               "date": "Thu, 01 Mar 2018 23:30:59 GMT",
               "x-amzn-requestid": "98d2c109-1da8-11e8-a2d9-b91cf22c7f33",
               "content-length": "449",
               "content-type": "application/x-amz-json-1.1",
               "connection": "keep-alive"
            }
         },
         "OrientationCorrection": "ROTATE_0"
      }
   }
]

Answer 1

big_json_file = json.loads(file_string)
big_list_of_labels = []

for file in big_json_file:
    big_list_of_labels.append(file['Analysis']['Labels'])

或者，如果您想存儲文件名和列表，我建議您使用以下內容：

my_processed_dict = {}
for file in big_json_file:
    my_processed_dict[file['File_Name']] = file['Analysis']['Labels']

您可以在其中迭代 my_processed_dict ：

for key, value in my_processed_dict.items():
    # value is the list of confidence values!
    pass