pandas AttributeError: 'Series' 对象没有属性 'as_matrix' 为什么会出错？
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/60164560/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
AttributeError: 'Series' object has no attribute 'as_matrix' Why is it error?
提问by YC Sang
When I execute the code of the official website, I get such an error. Why? code show as follow:
当我执行官网的代码时，出现这样的错误。为什么？代码显示如下：
landmarks_frame = pd.read_csv(‘F:\OfficialData\faces\face_landmarks.csv')
n = 65
img_name = landmarks_frame.iloc[n, 0]
landmarks = landmarks_frame.iloc[n, 1:].as_matrix()
landmarks = landmarks.astype(‘float').reshape(-1, 2)
List item
项目清单
回答by Sal Borrelli
As stated in another answer, the as_matrixmethod is deprecated since 0.23.0, so you should use to_numpyinstead. However, I want to highlight the fact that as_matrixand to_numpyhave different signatures: as_matrixtakes a list of column names as one of its parameter, in case you want to limit the conversion to a subset of the original DataFrame; to_numpydoes not accept such a parameter. As a consequence, the two methods are completely interchangeable only if you want to convert the DataFrame in full. If you (as in my case) need to convert a subset of the matrix, the usage would be quite different in the two use cases.
如另一个答案所述，该as_matrix方法自 0.23.0 起已弃用，因此您应该to_numpy改用。但是，我想强调一个事实，as_matrix并且to_numpy具有不同的签名：as_matrix将列名列表作为其参数之一，以防您想将转换限制为原始 DataFrame 的子集；to_numpy不接受这样的参数。因此，只有当您想完全转换 DataFrame 时，这两种方法才完全可以互换。如果您（如我的情况）需要转换矩阵的子集，则两种用例的用法将大不相同。
For example let's assume we only need to convert the subset ['col1', 'col2', 'col4'] of our original DataFrame to a Numpy array. In that case you might have some legacy code relying on as_matrixto convert, which looks more or less like:
例如，假设我们只需要将原始 DataFrame 的子集 ['col1', 'col2', 'col4'] 转换为 Numpy 数组。在这种情况下，您可能有一些依赖于as_matrix转换的遗留代码，它或多或少看起来像：
df.as_matrix(['col1', 'col2', 'col4'])
While converting the above code to to_numpyyou cannot simply replace the function name like in:
将上述代码转换为to_numpy您不能简单地替换函数名称，如：
df.to_numpy(['col1', 'col2', 'col4']) # WRONG
because to_numpydoes not accept a subset of columns as parameter. The solution in that case would be to do the selection first, and apply to_numpyto the result, as in:
因为to_numpy不接受列的子集作为参数。在这种情况下，解决方案是先进行选择，然后应用于to_numpy结果，如下所示：
df[['col1', 'col2', 'col4']].to_numpy() # CORRECT
回答by Remis Haroon
The purpose of as_matrixmethod is to
as_matrix方法的目的是
Convert the frame to its Numpy-array representation.
将帧转换为其 Numpy 数组表示。
as_matrixmethod is deprecated since 0.23.0
0.25.1 documentationsays : Deprecated since version 0.23.0: Use DataFrame.values() instead
as_matrix方法自 0.23.0
起已弃用0.25.1 文档说：自 0.23.0 版起已弃用：改为使用 DataFrame.values()
The two alternatives are
两种选择是
- .values() : Returns numpy.ndarray
- .to_numpy() : Returns numpy.ndarray
- .values() ：返回 numpy.ndarray
- .to_numpy() : 返回 numpy.ndarray
However, .values()documentationgives another warning :- Warning We recommend using DataFrame.to_numpy() instead.
但是，.values()文档给出了另一个警告：-Warning We recommend using DataFrame.to_numpy() instead.
I got the error in a slightly different way : AttributeError: 'DataFrame' object has no attribute 'as_matrix'
我以稍微不同的方式得到错误： AttributeError: 'DataFrame' object has no attribute 'as_matrix'
