【发布时间】:2018-08-05 11:54:23
【问题描述】:
我发现numpy.vectorize 允许将期望单个数字作为输入的“普通”函数转换为函数,该函数还可以将输入列表转换为函数已映射到每个函数的列表输入。例如,以下测试通过:
import numpy as np
import pytest
@np.vectorize
def f(x):
if x == 0:
return 1
else:
return 2
def test_1():
assert list(f([0, 1, 2])) == [1, 2, 2]
def test_2():
assert f(0) == 1
if __name__ == "__main__":
pytest.main([__file__])
但是,对于使用实例属性的实例方法,我无法让它工作。例如:
class Dummy(object):
def __init__(self, val=1):
self.val = val
@np.vectorize
def f(self, x):
if x == 0:
return self.val
else:
return 2
def test_3():
assert list(Dummy().f([0, 1, 2])) == [1, 2, 2]
此测试失败:
=================================== FAILURES ===================================
____________________________________ test_3 ____________________________________
def test_3():
> assert list(Dummy().f([0, 1, 2])) == [1, 2, 2]
test_numpy_vectorize.py:31:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/numpy/lib/function_base.py:2739: in __call__
return self._vectorize_call(func=func, args=vargs)
/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/numpy/lib/function_base.py:2809: in _vectorize_call
ufunc, otypes = self._get_ufunc_and_otypes(func=func, args=args)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
self = <numpy.lib.function_base.vectorize object at 0x106546470>
func = <function Dummy.f at 0x10653a2f0>, args = [array([0, 1, 2])]
def _get_ufunc_and_otypes(self, func, args):
"""Return (ufunc, otypes)."""
# frompyfunc will fail if args is empty
if not args:
raise ValueError('args can not be empty')
if self.otypes is not None:
otypes = self.otypes
nout = len(otypes)
# Note logic here: We only *use* self._ufunc if func is self.pyfunc
# even though we set self._ufunc regardless.
if func is self.pyfunc and self._ufunc is not None:
ufunc = self._ufunc
else:
ufunc = self._ufunc = frompyfunc(func, len(args), nout)
else:
# Get number of outputs and output types by calling the function on
# the first entries of args. We also cache the result to prevent
# the subsequent call when the ufunc is evaluated.
# Assumes that ufunc first evaluates the 0th elements in the input
# arrays (the input values are not checked to ensure this)
args = [asarray(arg) for arg in args]
if builtins.any(arg.size == 0 for arg in args):
raise ValueError('cannot call `vectorize` on size 0 inputs '
'unless `otypes` is set')
inputs = [arg.flat[0] for arg in args]
> outputs = func(*inputs)
E TypeError: f() missing 1 required positional argument: 'x'
是否可以将numpy.vectorize 应用于实例方法?
【问题讨论】:
-
装饰器版本
@vectorize比函数版本更难应用。您的方法f有两个参数,self和x。vectorize不够聪明,无法将Dummy()分配给self,并在[0, 1, 2]上进行迭代。 -
如果您打算对
np.vectorize做任何严肃的事情,我建议您研究它的代码。和np.frompyfunc一起玩。
标签: python numpy vectorization python-decorators