ValueError: 数据必须是一维的，而不是形状为 (6, 1) 的 ndarray。

huangapple
go评论210阅读模式
英文:

ValueError: Data must be 1-dimensional, got ndarray of shape (6, 1) instead

问题

我想使用索引、X和y变量数据创建一个新的数据框。

  1. df_idx1 = [[3],
  2.  [4],
  3.  [5],
  4.  [6],
  5.  [7],
  6.  [8]]
  7. X1 = [[[10],
  8.   [20],
  9.   [30]],
  11.  [[20],
  12.   [30],
  13.   [40]],
  15.  [[30],
  16.   [40],
  17.   [50]],
  19.  [[40],
  20.   [50],
  21.   [60]],
  23.  [[50],
  24.   [60],
  25.   [70]],
  27.  [[60],
  28.   [70],
  29.   [80]]]
  30. y1 = [[[40]],
  32.  [[50]],
  34.  [[60]],
  36.  [[70]],
  38.  [[80]],
  40.  [[90]]]
  42. print("Length index, X, Y: ", len(df_idx1), len(X1), len(y1))
  43. print("df_idx1", df_idx1)
  44. print("X1", X1)
  45. print("y1", y1)
  46. exdf1 = pd.DataFrame(data={"X": np.array(X1), "y": np.array(y1)}, index=df_idx1)

输出结果：

  1. exdf1=
  2.               X           y
  3. 3   [[10],[20],[30]]   [[40]]
  4. 4   [[20],[30],[40]]   [[50]]
  5. 5   [[30],[40],[50]]   [[60]]
  6. 6   [[40],[50],[60]]   [[70]]
  7. 7   [[50],[60],[70]]   [[80]]
  8. 8   [[60],[70],[80]]   [[90]]
英文:

I wanted to create a new dataframe using index, X, y variable data.

  1. df_idx1 = [[3]
  2.  [4]
  3.  [5]
  4.  [6]
  5.  [7]
  6.  [8]]
  7. X1 = [[[10]
  8.   [20]
  9.   [30]]
  11.  [[20]
  12.   [30]
  13.   [40]]
  15.  [[30]
  16.   [40]
  17.   [50]]
  19.  [[40]
  20.   [50]
  21.   [60]]
  23.  [[50]
  24.   [60]
  25.   [70]]
  27.  [[60]
  28.   [70]
  29.   [80]]]
  30. y1 = [[[40]]
  32.  [[50]]
  34.  [[60]]
  36.  [[70]]
  38.  [[80]]
  40.  [[90]]]
  42. print("Length index, X, Y: ", len(df_idx1), len(X1), len(y1))
  43. print("df_idx1",df_idx1)
  44. print("X1",X1)
  45. print("y1",y1)
  46. exdf1 = pd.DataFrame(data={"X":np.array(X1),"y":np.array(y1)},index=df_idx1)

present output:

  1. Length of index, X, Y:  6 6 6
  3. ValueError: Data must be 1-dimensional, got ndarray of shape (6, 1) instead

Expected output:

  1. exdf1=
  2.              X                y
  3. 3    [[10],[20],[30]]       [[40]]
  4. 4    [[20],[30],[40]]       [[50]]
  5. 5     ....
  6. 6
  7. 7
  8. 8    [[60],[70],[80]]       [[90]]

答案1

得分: 3

尝试这样做：

  1. idx = np.array(df_idx1).reshape(-1)
  3. df_X1 = pd.DataFrame(np.array(X1).reshape(6, -1))
  4. df_Y1 = pd.DataFrame(np.array(y1).reshape(-1))
  6. df_combined = pd.concat([x, y], axis=1)
  7. df_combined.set_index(idx, inplace=True)

结果将如下所示：
图片
(也许您可能想要重置列)

注意：
数据必须首先以正确的列表格式转换为nd.array

英文:

The question is not explicit and clear here, but I think you just want the error to be handled, which is making the data 1-dimensional and becoming able to make them as a DataFrame.

try this:

  1. idx = np.array(df_idx1).reshape(-1)
  3. df_X1 = pd.DataFrame(np.array(X1).reshape(6, -1))
  4. df_Y1 = pd.DataFrame(np.array(y1).reshape(-1))
  6. df_combined = pd.concat([x,y], axis=1)
  7. df_combined.set_index(idx, inplace=True)

The result would be like this
picture
(perhaps you might want to reset the columns)

Note:
the data must be in the right format of list to make them as an nd.array first

答案2

得分: 3

以下是您提供的代码的翻译：

使用像您的这样的数组：

  1. In [90]: idx = np.arange(3, 9).reshape(6, 1)
  2. In [91]: X = np.arange(10, 28).reshape(6, 3, 1); Y = 10 * np.arange(4, 10).reshape(6, 1, 1)

使用 idx 制作一个DataFrame会产生错误：

  1. In [92]: df = pd.DataFrame(index=idx, columns=['X', 'Y'])

Ravel的idx：

  1. In [93]: df = pd.DataFrame(index=idx.ravel(), columns=['X', 'Y'])

现在分配这两个系列：

  1. In [95]: df['X'] = list(X)
  2. In [96]: df['Y'] = list(Y)

数组提取自DataFrame可能有助于更好地理解实际存储的内容：

  1. In [111]: df.to_numpy()

提取 'X' 列的数组：

  1. In [112]: df['X'].to_numpy()

请注意，这些翻译中包含了代码的关键部分，不包括问题或其他内容。

英文:

With arrays like yours:

  1. In [90]: idx=np.arange(3,9).reshape(6,1)    
  2. In [91]: X = np.arange(10,28).reshape(6,3,1); Y = 10*np.arange(4,10).reshape(6,1,1)

Making a frame with idx produces your error:

  1. In [92]: df=pd.DataFrame( index=idx, columns=['X','Y'])
  2. ---------------------------------------------------------------------------
  3. ValueError                                Traceback (most recent call last)
  4. /tmp/ipykernel_7567/2068536625.py in <module>
  5. ----> 1 df=pd.DataFrame( index=idx, columns=['X','Y'])
  7. ~/.local/lib/python3.10/site-packages/pandas/core/frame.py in __init__(self, data, index, columns, dtype, copy)
  8.     662         elif isinstance(data, dict):
  9.     663             # GH#38939 de facto copy defaults to False only in non-dict cases
  10. --> 664             mgr = dict_to_mgr(data, index, columns, dtype=dtype, copy=copy, typ=manager)
  11.     665         elif isinstance(data, ma.MaskedArray):
  12.     666             import numpy.ma.mrecords as mrecords
  14. ~/.local/lib/python3.10/site-packages/pandas/core/internals/construction.py in dict_to_mgr(data, index, columns, dtype, typ, copy)
  15.     448             index = _extract_index(arrays[~missing])
  16.     449         else:
  17. --> 450             index = ensure_index(index)
  18.     451 
  19.     452         # no obvious "empty" int column
  21. ~/.local/lib/python3.10/site-packages/pandas/core/indexes/base.py in ensure_index(index_like, copy)
  22.    7331             return Index._with_infer(index_like, copy=copy, tupleize_cols=False)
  23.    7332     else:
  24. -> 7333         return Index._with_infer(index_like, copy=copy)
  25.    7334 
  26.    7335 
  28. ~/.local/lib/python3.10/site-packages/pandas/core/indexes/base.py in _with_infer(cls, *args, **kwargs)
  29.     714         with warnings.catch_warnings():
  30.     715             warnings.filterwarnings("ignore", ".*the Index constructor", FutureWarning)
  31. --> 716             result = cls(*args, **kwargs)
  32.     717 
  33.     718         if result.dtype == _dtype_obj and not result._is_multi:
  35. ~/.local/lib/python3.10/site-packages/pandas/core/indexes/base.py in __new__(cls, data, dtype, copy, name, tupleize_cols, **kwargs)
  36.     538 
  37.     539             klass = cls._dtype_to_subclass(arr.dtype)
  38. --> 540             arr = klass._ensure_array(arr, dtype, copy)
  39.     541             disallow_kwargs(kwargs)
  40.     542             return klass._simple_new(arr, name)
  42. ~/.local/lib/python3.10/site-packages/pandas/core/indexes/numeric.py in _ensure_array(cls, data, dtype, copy)
  43.     174         if subarr.ndim > 1:
  44.     175             # GH#13601, GH#20285, GH#27125
  45. --> 176             raise ValueError("Index data must be 1-dimensional")
  46.     177 
  47.     178         subarr = np.asarray(subarr)
  49. ValueError: Index data must be 1-dimensional

Ravel of idx:

  1. In [93]: df=pd.DataFrame( index=idx.ravel(), columns=['X','Y'])
  3. In [94]: df
  4. Out[94]: 
  5.      X    Y
  6. 3  NaN  NaN
  7. 4  NaN  NaN
  8. 5  NaN  NaN
  9. 6  NaN  NaN
  10. 7  NaN  NaN
  11. 8  NaN  NaN

Now assign the 2 series:

  1. In [95]: df['X']=list(X)    
  2. In [96]: df['Y']=list(Y)
  4. In [97]: df
  5. Out[97]: 
  6.                     X       Y
  7. 3  [[10], [11], [12]]  [[40]]
  8. 4  [[13], [14], [15]]  [[50]]
  9. 5  [[16], [17], [18]]  [[60]]
  10. 6  [[19], [20], [21]]  [[70]]
  11. 7  [[22], [23], [24]]  [[80]]
  12. 8  [[25], [26], [27]]  [[90]]

I tried various things using data=... parameter, but kept getting various errors, mainly a conflict between the implied columns of X and Y and the desired two. And the list() was also needed - each Series is object dtype, with 6 separate arrays.

edit

The array extracted from the dataframe may help you understand better what is actually being stored.

  1. In [111]: df.to_numpy()
  2. Out[111]: 
  3. array([[array([[10],
  4.                [11],
  5.                [12]]), array([[40]])],
  6.        [array([[13],
  7.                [14],
  8.                [15]]), array([[50]])],
  9.        [array([[16],
  10.                [17],
  11.                [18]]), array([[60]])],
  12.        [array([[19],
  13.                [20],
  14.                [21]]), array([[70]])],
  15.        [array([[22],
  16.                [23],
  17.                [24]]), array([[80]])],
  18.        [array([[25],
  19.                [26],
  20.                [27]]), array([[90]])]], dtype=object)
  22. In [112]: df['X'].to_numpy()
  23. Out[112]: 
  24. array([array([[10],
  25.               [11],
  26.               [12]]), array([[13],
  27.                              [14],
  28.                              [15]]), array([[16],
  29.                                             [17],
  30.                                             [18]]), array([[19],
  31.                                                            [20],
  32.                                                            [21]]),
  33.        array([[22],
  34.               [23],
  35.               [24]]), array([[25],
  36.                              [26],
  37.                              [27]])], dtype=object)

huangapple
  • 本文由 发表于 2023年6月8日 09:23:48
  • 转载请务必保留本文链接：https://go.coder-hub.com/76428038.html
  • dataframe
  • numpy
  • pandas
  • python
  • python-3.x
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定