I have two Anaconda environments:
- The base environment
- An R environment (containing R only)
The base environment is the default environment and does not appear in the Anaconda subdirectory envs. Base is activated by default and in particular while the commands below are executed.
When I run python from within the Anaconda prompt (or another prompt that I have configured for Anaconda), importing numpy works without problems.
However, when I run my scrips from my IDE (exctuing C:\my\path\to\Anaconda3\python.exe myscript.py), the import of numpy fails. Therefore, I have installed numpy through pip and ended up with a broken installation. import numpy results now in
Traceback (most recent call last):
File "D:\GoogleDrive\_Backup\Programmieren\Datathlon\econ_model.py", line 1, in <module>
import numpy as np
File "C:\ProgramData\Anaconda3\lib\site-packages\numpy\__init__.py", line 140, in <module>
from . import _distributor_init
File "C:\ProgramData\Anaconda3\lib\site-packages\numpy\_distributor_init.py", line 34, in <module>
from . import _mklinit
ImportError: DLL load failed: Das angegebene Modul wurde nicht gefunden.
As far as I know the issue is due to different versions of numpy being used. (I want the mkl-enabled version, though).
That the right version of numpy is not found lets me guess that I have multiple python environments somehow and that the wrong one is called by my IDE. I have no idea how this can be possible, as Anaconda is the only python resource on my PC.
How can I call the right python version from outside of the anaconda prompt? How can I configure python/Anaconda that the correct version is called by default?
I am aware that activating the base environment solves the problem when working within a shell. But if I run a single command, i.e. just call the python script without opening a shell before, I cannot activate anything (can I?).
Techincal details:
- Python 3.7
- conda version 4.6.8 win-64
- Windows 10 64bit
- IDE: Eclipse with PyDev