英文:
numpy stack not working for astropy quantities in numpy 1.23
问题
以下是翻译好的部分：
在numpy 1.23.5中以下代码可行，但在1.24.3中不行：
from astropy import units as u
import numpy as np
a = np.array([[1, 2], [3, 4]])
b = np.array([[5, 6], [7, 8]])
np.stack([a,b]*u.m)
使用numpy 1.24.3时，我得到以下错误：
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "<__array_function__ internals>", line 200, in stack
File "~/opt/anaconda3/lib/python3.9/site-packages/astropy/units/quantity.py", line 1683, in __array_function__
return super().__array_function__(function, types, args, kwargs)
File "~/opt/anaconda3/lib/python3.9/site-packages/numpy/core/shape_base.py", line 471, in stack
return _nx.concatenate(expanded_arrays, axis=axis, out=out,
File "<__array_function__ internals>", line 200, in concatenate
File "~/opt/anaconda3/lib/python3.9/site-packages/astropy/units/quantity.py", line 1688, in __array_function__
args, kwargs, unit, out = function_helper(*args, **kwargs)
TypeError: concatenate() got an unexpected keyword argument 'dtype'
而在numpy 1.23.5中，我得到：
<Quantity [[[1., 2.],
[3., 4.]],
[[5., 6.],
[7., 8.]]] m>
我是否做错了什么，或者是numpy 1.24引入了一个错误，还是需要更新astropy units以使其与新的numpy版本兼容？
英文:
The following works in numpy 1.23.5 but not in 1.24.3:
from astropy import units as u
import numpy as np
a = np.array([[1, 2], [3, 4]])
b = np.array([[5, 6], [7, 8]])
np.stack([a,b]*u.m)
With numpy 1.24.3, I get:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "<__array_function__ internals>", line 200, in stack
File "~/opt/anaconda3/lib/python3.9/site-packages/astropy/units/quantity.py", line 1683, in __array_function__
return super().__array_function__(function, types, args, kwargs)
File "~/opt/anaconda3/lib/python3.9/site-packages/numpy/core/shape_base.py", line 471, in stack
return _nx.concatenate(expanded_arrays, axis=axis, out=out,
File "<__array_function__ internals>", line 200, in concatenate
File "~/opt/anaconda3/lib/python3.9/site-packages/astropy/units/quantity.py", line 1688, in __array_function__
args, kwargs, unit, out = function_helper(*args, **kwargs)
TypeError: concatenate() got an unexpected keyword argument 'dtype'
While with numpy 1.23.5, I get:
<Quantity [[[1., 2.],
[3., 4.]],
[[5., 6.],
[7., 8.]]] m>
Am I doing something wrong or has been a bug introduced in numpy 1.24 or does astropy units have to be updated to work with the new numpy version?
答案1
得分: 2
这是我的设置，对我有效：
Python 3.9.12（主要版本，2022年4月5日，01:53:17）[Clang 12.0.0]；
Numpy 1.24.3；
Astropy 5.3。
在NumPy 1.24.3版本中，开发人员为函数stack添加了新的参数：casting和dtype。
根据提供的错误消息，看起来astropy收到了意外的关键字dtype，这可能是一个问题。我假设您的astropy版本为5.1，并建议升级至5.3。
英文:
It's work for me with setup:
Python 3.9.12 (main, Apr 5 2022, 01:53:17) [Clang 12.0.0 ]<br>
Numpy 1.24.3<br>
Astropy 5.3
In version NumPy 1.24.3 developers added new arguments for function stack: <i>casting</i> and <i>dtype</i>.
As I see in provided error message, astropy got unexpected keyword dtype and that may be an issue here. I assume that your version of astropy - 5.1 and recommend upgrade to 5.3.
答案2
得分: 0
遇到了相同的消息/问题 - Conda 无法使用 "conda update astropy" 更新到 5.3.1 版本的 astropy（一直给我 v5.1）。最后我使用 pip install 安装了它。即使使用 astropy 5.3.1，它仍然无法与 numpy 1.24.3 一起工作。不得不降级到 numpy 1.23.5，使用 "conda install numpy=1.23.5"。现在一切都正常。
英文:
Had the same message/issue - Conda wouldn't update astropy to 5.3.1 using "conda update astropy" (kept giving me v5.1). Ended up doing a pip install. Even with astropy 5.3.1 it still wouldn't work with numpy 1.24.3. Had to downgrade to numpy 1.23.5. doing "conda install numpy=1.23.5". Now all is well.
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论