Conda示例不起作用，它显示的是较新而不是较旧的numpy？

提问者：小点点

Conda示例不起作用，它显示的是较新而不是较旧的numpy？

我从这里运行这个例子；http://conda.pydata.org/docs/index.html

我在苹果电脑上，10.8.5。我可能肯定压扁了一些东西，因为我最初对我的蟒蛇有问题，但我认为它现在起作用了。为了进行理智检查，我运行了上面帮助中显示的示例。它只是创建了一个新的conda环境，使用了一个较旧的Numpy版本，然后显示了源激活[env]如何获得首选项。

就我而言，一切似乎都很顺利，但numpy的版本保持不变。有什么想法吗？提前非常感谢！

$ ~/anaconda/bin/python2.7
Python 2.7.6 |Anaconda 1.9.2 (x86_64)| (default, Jan 10 2014, 11:23:15)
[GCC 4.0.1 (Apple Inc. build 5493)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import numpy
nu>>> numpy.__version__
'1.8.1'
>>> quit()

$ conda create -p ~/anaconda/envs/trynum16 numpy=1.6 anaconda

$ source activate ~/anaconda/envs/trynum16
discarding ~/anaconda/bin from PATH
prepending ~/anaconda/envs/trynum16/bin to PATH

(~/anaconda/envs/trynum16):~ $ echo $PATH
~/anaconda/envs/trynum16/bin:/usr/local/bin:/Library/Frameworks/Python.framework/Versions/2.7/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/opt/X11/bin:/usr/local/git/bin:/bin

(~/anaconda/envs/trynum16):~ $ python
Python 2.7.5 |Anaconda 1.8.0 (x86_64)| (default, Oct 24 2013, 07:02:20)
[GCC 4.0.1 (Apple Inc. build 5493)] on darwin
Type "help", "copyright", "credits" or "license" for more information.

>>> import numpy
>>> numpy.__version__
'1.8.1'
>>> quit()

(~/anaconda/envs/trynum16):~ $ echo $PATH
~/anaconda/envs/trynum16/bin:/usr/local/bin:/Library/Frameworks/Python.framework/Versions/2.7/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/opt/X11/bin:/usr/local/git/bin:/bin

(~/anaconda/envs/trynum16):~ $ source deactivate
discarding ~/anaconda/envs/trynum16/bin from PATH

$ echo $PATH
~/anaconda/bin:/usr/local/bin:/Library/Frameworks/Python.framework/Versions/2.7/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/opt/X11/bin:/usr/local/git/bin:/bin

共1个答案

匿名用户

OK，这是用户错误。我的顶级蟒蛇，非蟒蛇，有一个Numpy，它首先被捡起来！为了确定冲突并解决它，我做了以下工作。

# Start from Non-Anaconda or Root Python environment
$ source deactivate
$ conda info -e
# conda environments:
trynum16                 ~/anaconda/envs/trynum16
root                  *  ~/anaconda

$ which python
/usr/local/bin/python

$ echo $PATH
/usr/local/bin:~/anaconda/bin:/Library/Frameworks/Python.framework/Versions/2.7/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/opt/X11/bin:/usr/local/git/bin:/bin

# Amongst many other items, see the 'offending' numpy here
$ pip freeze
numpy==1.8.1

# Without Sudo, got error,
$ pip uninstall numpy
OSError: [Errno 13] Permission denied: '/Library/Python/2.7/site-packages/bson/__init__.py'
# So did Sudo, probably my mistake on the install?
$ sudo pip uninstall numpy

$ source activate ~/anaconda/envs/trynum16
$ conda info -e
# conda environments:
trynum16              * ~/anaconda/envs/trynum16
root                    ~/anaconda
$ python
Python 2.7.6 |Continuum Analytics, Inc.| (default, Jan 10 2014, 11:23:15)
[GCC 4.0.1 (Apple Inc. build 5493)] on darwin
>>> import numpy
>>> numpy.__version__
'1.6.2'

完成此操作后，两个不同版本的独立conda环境可以正常工作。下面是如何完成第二步的。

预检查：1。查看PYPI上存在的Numpy版本：https://pypi.python.org/pypi/numpy.
该页面仅显示最新版本，因此请在此处查找历史项目；下载网址：http://sourceforge.net/projects/numpy/files/NumPy/

2.查看Binstar上的Numpy版本；https://binstar.org/search?q=numpy

# 3.Which NumPy versions have official, public Conda packages?
$ conda search numpy
Fetching package metadata: ...
numpy                        1.5.1                    py27_0  defaults
#...etc

# This old version Conflicted with Anaconda, so removed that part.
$ conda create -p ~/anaconda/envs/trynum15 numpy=1.5.1
$ source activate /Users/amoroney/anaconda/envs/trynum15
$ conda info -e
# conda environments:
trynum15              *  ~/anaconda/envs/trynum15
trynum16                 ~/anaconda/envs/trynum16
root                     ~/anaconda

$ python
>>> import numpy
>>> numpy.__version__
'1.5.1'