您好我遇到了json字符串到data.frame转换的问题。
print (json_resp.text)
{
"error_code": 0,
"description": "",
"img_size": { "w": 650, "h": 488 },
"people": [
{
"age": 22,
"gender": 84,
"mood": 29,
"position": { "x": 190, "y": 161, "w": 259, "h": 259 },
"rotation": { "yaw": -3, "pitch": 3, "roll": -1 },
"landmarks": { "lefteye": { "x": 371, "y": 233 }, "righteye": { "x": 266, "y": 236 }, "maskpoints": [ { "x": 371, "y": 233 }, { "x": 266, "y": 236 }, { "x": 203, "y": 234 }, { "x": 206, "y": 261 }, { "x": 212, "y": 287 }, { "x": 220, "y": 313 }, { "x": 233, "y": 338 }, { "x": 250, "y": 357 }, { "x": 273, "y": 373 }, { "x": 296, "y": 388 }, { "x": 321, "y": 394 }, { "x": 346, "y": 390 }, { "x": 371, "y": 377 }, { "x": 396, "y": 362 }, { "x": 416, "y": 341 }, { "x": 430, "y": 315 }, { "x": 437, "y": 287 }, { "x": 444, "y": 258 }, { "x": 448, "y": 227 }, { "x": 215, "y": 221 }, { "x": 226, "y": 203 }, { "x": 247, "y": 195 }, { "x": 269, "y": 198 }, { "x": 291, "y": 204 }, { "x": 336, "y": 201 }, { "x": 360, "y": 193 }, { "x": 385, "y": 190 }, { "x": 408, "y": 198 }, { "x": 423, "y": 216 }, { "x": 314, "y": 231 }, { "x": 314, "y": 247 }, { "x": 314, "y": 263 }, { "x": 314, "y": 279 }, { "x": 294, "y": 297 }, { "x": 304, "y": 300 }, { "x": 315, "y": 302 }, { "x": 327, "y": 299 }, { "x": 338, "y": 297 }, { "x": 242, "y": 236 }, { "x": 254, "y": 226 }, { "x": 271, "y": 226 }, { "x": 284, "y": 240 }, { "x": 270, "y": 245 }, { "x": 253, "y": 244 }, { "x": 349, "y": 238 }, { "x": 362, "y": 224 }, { "x": 379, "y": 224 }, { "x": 393, "y": 234 }, { "x": 381, "y": 242 }, { "x": 363, "y": 242 }, { "x": 281, "y": 332 }, { "x": 294, "y": 327 }, { "x": 306, "y": 322 }, { "x": 315, "y": 325 }, { "x": 325, "y": 323 }, { "x": 340, "y": 328 }, { "x": 357, "y": 335 }, { "x": 341, "y": 347 }, { "x": 327, "y": 354 }, { "x": 317, "y": 354 }, { "x": 306, "y": 353 }, { "x": 294, "y": 347 }, { "x": 289, "y": 333 }, { "x": 306, "y": 331 }, { "x": 316, "y": 332 }, { "x": 325, "y": 331 }, { "x": 349, "y": 334 }, { "x": 326, "y": 339 }, { "x": 316, "y": 340 }, { "x": 306, "y": 339 } ] },
"clothingcolors": [ ],
"ethnicity": { "african": 83, "asian": 0, "caucasian": 12, "hispanic": 3 },
"emotions": { "happiness": 1, "surprise": 5, "anger": 2, "disgust": 2, "fear": 1, "sadness": 11 }
}
]
但是，当我尝试将json字符串更改为data.frame时，我得到：
import pandas as pd
df_json = pd.read_json(json_resp.text, typ='frame')
print (df_json)
Traceback (most recent call last):
File "<input>", line 1, in <module>
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\io\json\json.py", line 427, in read_json
result = json_reader.read()
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\io\json\json.py", line 537, in read
obj = self._get_object_parser(self.data)
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\io\json\json.py", line 556, in _get_object_parser
obj = FrameParser(json, **kwargs).parse()
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\io\json\json.py", line 652, in parse
self._parse_no_numpy()
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\io\json\json.py", line 871, in _parse_no_numpy
loads(json, precise_float=self.precise_float), dtype=None)
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\core\frame.py", line 392, in __init__
mgr = init_dict(data, index, columns, dtype=dtype)
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\core\internals\construction.py", line 212, in init_dict
return arrays_to_mgr(arrays, data_names, index, columns, dtype=dtype)
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\core\internals\construction.py", line 51, in arrays_to_mgr
index = extract_index(arrays)
File "C:\Users\uzytkownik\PycharmProjects\Face API\venv\lib\site-packages\pandas\core\internals\construction.py", line 320, in extract_index
raise ValueError('Mixing dicts with non-Series may lead to '
ValueError: Mixing dicts with non-Series may lead to ambiguous ordering.
我应该在代码中更改什么才能获得简单的data.frame？
json_normalize是你想要做的。但是，有嵌套列表，这意味着它只能标准化/展平到第一级。
我认为问题来自于landmarks.maskpoints，因为它创建了70行，其中包含2列x和y。因此，尝试创建单行，包含70行的内容可能是个问题。
如果你只是一点一点地开始尝试打开/压扁它，你可以看看我的意思。基本上要展平，你想要标准化每个部分，然后在最后将它们合并成一行，但你可以看到掩码点的问题。
jsonStr = '''
{
"error_code": 0,
"description": "",
"img_size": { "w": 650, "h": 488 },
"people": [
{
"age": 22,
"gender": 84,
"mood": 29,
"position": { "x": 190, "y": 161, "w": 259, "h": 259 },
"rotation": { "yaw": -3, "pitch": 3, "roll": -1 },
"landmarks": { "lefteye": { "x": 371, "y": 233 }, "righteye": { "x": 266, "y": 236 }, "maskpoints": [ { "x": 371, "y": 233 }, { "x": 266, "y": 236 }, { "x": 203, "y": 234 }, { "x": 206, "y": 261 }, { "x": 212, "y": 287 }, { "x": 220, "y": 313 }, { "x": 233, "y": 338 }, { "x": 250, "y": 357 }, { "x": 273, "y": 373 }, { "x": 296, "y": 388 }, { "x": 321, "y": 394 }, { "x": 346, "y": 390 }, { "x": 371, "y": 377 }, { "x": 396, "y": 362 }, { "x": 416, "y": 341 }, { "x": 430, "y": 315 }, { "x": 437, "y": 287 }, { "x": 444, "y": 258 }, { "x": 448, "y": 227 }, { "x": 215, "y": 221 }, { "x": 226, "y": 203 }, { "x": 247, "y": 195 }, { "x": 269, "y": 198 }, { "x": 291, "y": 204 }, { "x": 336, "y": 201 }, { "x": 360, "y": 193 }, { "x": 385, "y": 190 }, { "x": 408, "y": 198 }, { "x": 423, "y": 216 }, { "x": 314, "y": 231 }, { "x": 314, "y": 247 }, { "x": 314, "y": 263 }, { "x": 314, "y": 279 }, { "x": 294, "y": 297 }, { "x": 304, "y": 300 }, { "x": 315, "y": 302 }, { "x": 327, "y": 299 }, { "x": 338, "y": 297 }, { "x": 242, "y": 236 }, { "x": 254, "y": 226 }, { "x": 271, "y": 226 }, { "x": 284, "y": 240 }, { "x": 270, "y": 245 }, { "x": 253, "y": 244 }, { "x": 349, "y": 238 }, { "x": 362, "y": 224 }, { "x": 379, "y": 224 }, { "x": 393, "y": 234 }, { "x": 381, "y": 242 }, { "x": 363, "y": 242 }, { "x": 281, "y": 332 }, { "x": 294, "y": 327 }, { "x": 306, "y": 322 }, { "x": 315, "y": 325 }, { "x": 325, "y": 323 }, { "x": 340, "y": 328 }, { "x": 357, "y": 335 }, { "x": 341, "y": 347 }, { "x": 327, "y": 354 }, { "x": 317, "y": 354 }, { "x": 306, "y": 353 }, { "x": 294, "y": 347 }, { "x": 289, "y": 333 }, { "x": 306, "y": 331 }, { "x": 316, "y": 332 }, { "x": 325, "y": 331 }, { "x": 349, "y": 334 }, { "x": 326, "y": 339 }, { "x": 316, "y": 340 }, { "x": 306, "y": 339 } ] },
"clothingcolors": [ ],
"ethnicity": { "african": 83, "asian": 0, "caucasian": 12, "hispanic": 3 },
"emotions": { "happiness": 1, "surprise": 5, "anger": 2, "disgust": 2, "fear": 1, "sadness": 11 }
}
]
}'''
import json
from pandas.io.json import json_normalize
jsonObj = json.loads(jsonStr)
# flatten at 1st level. But still nested lists/dictionaries in column `people`
df_a = json_normalize(jsonObj)
# so flatten out people, and you'll see clothingcolors still has a list and landmarks too
df_people = json_normalize(jsonObj['people'])
df_clothingcolors = json_normalize(jsonObj['people'][0]['clothingcolors'])
df_landmarks = json_normalize(jsonObj['people'][0]['landmarks'])
# the landmarks column still need to flatten maskpoints...but maskpoints produces 70 rows, and there's your issue
df_maskpoints = json_normalize(jsonObj['people'][0]['landmarks']['maskpoints'])
所以，如果你看看这些形状：
print (df_a.shape)
(1, 5)
print (df_people.shape)
(1, 26)
print (df_clothingcolors.shape)
(0, 0)
print (df_landmarks.shape)
(1, 5)
print (df_maskpoints.shape)
(70, 2)
...你看到maskpoints的形状是70行。
但，
我发现这个blog很有用。从本质上讲，它会解开所有嵌套列表，以便最终得到一个大的平面表。
jsonStr = '''
{
"error_code": 0,
"description": "",
"img_size": { "w": 650, "h": 488 },
"people": [
{
"age": 22,
"gender": 84,
"mood": 29,
"position": { "x": 190, "y": 161, "w": 259, "h": 259 },
"rotation": { "yaw": -3, "pitch": 3, "roll": -1 },
"landmarks": { "lefteye": { "x": 371, "y": 233 }, "righteye": { "x": 266, "y": 236 }, "maskpoints": [ { "x": 371, "y": 233 }, { "x": 266, "y": 236 }, { "x": 203, "y": 234 }, { "x": 206, "y": 261 }, { "x": 212, "y": 287 }, { "x": 220, "y": 313 }, { "x": 233, "y": 338 }, { "x": 250, "y": 357 }, { "x": 273, "y": 373 }, { "x": 296, "y": 388 }, { "x": 321, "y": 394 }, { "x": 346, "y": 390 }, { "x": 371, "y": 377 }, { "x": 396, "y": 362 }, { "x": 416, "y": 341 }, { "x": 430, "y": 315 }, { "x": 437, "y": 287 }, { "x": 444, "y": 258 }, { "x": 448, "y": 227 }, { "x": 215, "y": 221 }, { "x": 226, "y": 203 }, { "x": 247, "y": 195 }, { "x": 269, "y": 198 }, { "x": 291, "y": 204 }, { "x": 336, "y": 201 }, { "x": 360, "y": 193 }, { "x": 385, "y": 190 }, { "x": 408, "y": 198 }, { "x": 423, "y": 216 }, { "x": 314, "y": 231 }, { "x": 314, "y": 247 }, { "x": 314, "y": 263 }, { "x": 314, "y": 279 }, { "x": 294, "y": 297 }, { "x": 304, "y": 300 }, { "x": 315, "y": 302 }, { "x": 327, "y": 299 }, { "x": 338, "y": 297 }, { "x": 242, "y": 236 }, { "x": 254, "y": 226 }, { "x": 271, "y": 226 }, { "x": 284, "y": 240 }, { "x": 270, "y": 245 }, { "x": 253, "y": 244 }, { "x": 349, "y": 238 }, { "x": 362, "y": 224 }, { "x": 379, "y": 224 }, { "x": 393, "y": 234 }, { "x": 381, "y": 242 }, { "x": 363, "y": 242 }, { "x": 281, "y": 332 }, { "x": 294, "y": 327 }, { "x": 306, "y": 322 }, { "x": 315, "y": 325 }, { "x": 325, "y": 323 }, { "x": 340, "y": 328 }, { "x": 357, "y": 335 }, { "x": 341, "y": 347 }, { "x": 327, "y": 354 }, { "x": 317, "y": 354 }, { "x": 306, "y": 353 }, { "x": 294, "y": 347 }, { "x": 289, "y": 333 }, { "x": 306, "y": 331 }, { "x": 316, "y": 332 }, { "x": 325, "y": 331 }, { "x": 349, "y": 334 }, { "x": 326, "y": 339 }, { "x": 316, "y": 340 }, { "x": 306, "y": 339 } ] },
"clothingcolors": [ ],
"ethnicity": { "african": 83, "asian": 0, "caucasian": 12, "hispanic": 3 },
"emotions": { "happiness": 1, "surprise": 5, "anger": 2, "disgust": 2, "fear": 1, "sadness": 11 }
}
]
}'''
import json
from pandas.io.json import json_normalize
def flatten_json(y):
out = {}
def flatten(x, name=''):
if type(x) is dict:
for a in x:
flatten(x[a], name + a + '_')
elif type(x) is list:
i = 0
for a in x:
flatten(a, name + str(i) + '_')
i += 1
else:
out[name[:-1]] = x
flatten(y)
return out
jsonObj = json.loads(jsonStr)
flat = flatten_json(jsonObj)
df = json_normalize(flat)
输出将是您的1行，包含168列。