TypingError:在 nopython 模式管道中失败(步骤:nopython 前端) [英] TypingError: Failed in nopython mode pipeline (step: nopython frontend)

问题描述

我正在尝试使用 numba jit 编写我的第一个函数，我有一个 Pandas 数据框，我需要遍历它并找到每 350 个点的均方根，因为 python 的 for 循环非常慢，我决定尝试numba jit，代码是:

I am trying to write my first function using numba jit, I have a pandas dataframe that I need to iterate through and find the root mean square for each 350 points, since the for loop of python is quite slow I decided to try numba jit, the code is:

@jit(nopython=True)
def find_rms(data, length):
    res = []
    for i in range(length, len(data)):
        interval = np.array(data[i-length:i])
        interval =np.power(interval, 2)
        sum = interval.sum()
        resI = sum/length
        resI = np.sqrt(res)
        res.appennd(resI)
    return res

mydf = np.array(df.iloc[:]['c0'], dtype=np.float64)
df.iloc[350:]['rms'] = find_rms(mydf, 350)

我在某处读到我需要指定数据类型，因此我写了dtype = np.float64"但我仍然收到错误消息:

I read somewhere thad I need to specify datatypes, therefore I wrote "dtype = np.float64" but I still get the error as:

---------------------------------------------------------------------------
TypingError                               Traceback (most recent call last)
<ipython-input-39-4d388f72efdc> in <module>
----> 1 df.iloc[350:]['rms'] = find_rms(mydf, 350.0)

c:\users\1\appdata\local\programs\python\python35\lib\site-packages\numba\dispatcher.py in _compile_for_args(self, *args, **kws)
    346                 e.patch_message(msg)
    347 
--> 348             error_rewrite(e, 'typing')
    349         except errors.UnsupportedError as e:
    350             # Something unsupported is present in the user code, add help info

c:\users\1\appdata\local\programs\python\python35\lib\site-packages\numba\dispatcher.py in error_rewrite(e, issue_type)
    313                 raise e
    314             else:
--> 315                 reraise(type(e), e, None)
    316 
    317         argtypes = []

c:\users\1\appdata\local\programs\python\python35\lib\site-packages\numba\six.py in reraise(tp, value, tb)
    656             value = tp()
    657         if value.__traceback__ is not tb:
--> 658             raise value.with_traceback(tb)
    659         raise value
    660 

TypingError: Failed in nopython mode pipeline (step: nopython frontend)
Invalid use of Function(<built-in function array>) with argument(s) of type(s): (array(float64, 1d, C))
 * parameterized
In definition 0:
    TypingError: array(float64, 1d, C) not allowed in a homogeneous sequence
    raised from c:\users\1\appdata\local\programs\python\python35\lib\site-packages\numba\typing\npydecl.py:463
In definition 1:
    TypingError: array(float64, 1d, C) not allowed in a homogeneous sequence
    raised from c:\users\1\appdata\local\programs\python\python35\lib\site-packages\numba\typing\npydecl.py:463
This error is usually caused by passing an argument of a type that is unsupported by the named function.
[1] During: resolving callee type: Function(<built-in function array>)
[2] During: typing of call at <ipython-input-34-edd252715b2d> (5)


File "<ipython-input-34-edd252715b2d>", line 5:
def find_rms(data, length):
    <source elided>
    for i in range(length, len(data)):
        interval = np.array(data[i-length:i])
        ^

This is not usually a problem with Numba itself but instead often caused by
the use of unsupported features or an issue in resolving types.

To see Python/NumPy features supported by the latest release of Numba visit:
http://numba.pydata.org/numba-doc/dev/reference/pysupported.html
and
http://numba.pydata.org/numba-doc/dev/reference/numpysupported.html

For more information about typing errors and how to debug them visit:
http://numba.pydata.org/numba-doc/latest/user/troubleshoot.html#my-code-doesn-t-compile

If you think your code should work with Numba, please report the error message
and traceback, along with a minimal reproducer at:
https://github.com/numba/numba/issues/new

有人知道是什么问题吗?

Does anybody know what the problem is?

推荐答案

您在 append 中有一个错字，而且我认为您在取平方根时也犯了一个错误(我相信 resI 而不是 res).

You had a typo in append and I think you also made a mistake with what the square root is to be taken of (I believe resI not res).

除此之外，唯一的问题是interval的初始化.Numba 不希望您将 numpy 数组传递给 numpy 数组.将 np.array 包裹在数组的切片周围没有任何帮助，python 根本不在乎你是否这样做并且像对待代码一样对待代码，但是 nopython 模式下的 Numba 确实关心并抛出错误.去掉那部分就解决了问题.

Other than that, the only problem was the initialization of interval. Numba doesn't want you to pass a numpy array to a numpy array. It doesn't help with anything to wrap the np.array around the slice of the array, python simply doesn't care if you do that and treats the code like you didn't but Numba in nopython mode does care and throws an error. Leaving that part out solved the problem.

@jit(nopython=True)
def find_rms(data, length):
    res = []
    for i in range(length, len(data)):
        interval = data[i-length:i]
        interval = np.power(interval, 2)
        sum = interval.sum()
        resI = sum/length
        resI = np.sqrt(resI)
        res.append(resI)
    return res

mydf = np.array(df.iloc[:]['c0'], dtype=np.float64)
target = find_rms(mydf, 350)

这篇关于TypingError:在 nopython 模式管道中失败(步骤:nopython 前端)的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！