简体   繁体   中英

python: after installing anaconda, how to import pandas

I have installed anaconda. Now when i am trying to run

import pandas as pd

I am getting the following error

Traceback (most recent call last):
File "<pyshell#0>", line 1, in <module>
import pandasFile
ImportError: No module named pandasFile

It is my first day to python. I cannot figure out how to fix it. I am hoping that I have to change some path somewhere. I know it can be a silly question to post here.

I'm using python 3.4 and Anaconda3 4.2.

I had the same problem, but it worked (the import pandas works now anyway) for me to install pandas with pip by writing:

python -m pip install pandas

Good luck!

The cool thing about anaconda is, that you can manage virtual environments for several projects. Those also have the benefit of keeping several python installations apart. This could be a problem when several installations of a module or package are interfering with each other.

Try the following:

  1. Create a new anaconda environment with user@machine:~$ conda create -n pandas_env python=2.7
  2. Activate the environment with user@machine:~$ source activate pandas_env on Linux/OSX or $ activate pandas_env on Windows. On Linux the active environment is shown in parenthesis in front of the user name in the shell. (I am not sure how windows handles this, but you can see it by typing $ conda info -e . The one with the * next to it is the active one)
  3. Type (pandas_env)user@machine:~$ conda list to show a list of all installed modules.
  4. If pandas is missing from this list, install it (while still inside the pandas_env environment) with (pandas_env)user@machine:~$ conda install pandas , as @Fiabetto suggested.
  5. Open python (pandas_env)user@machine:~$ python and try to load pandas again.

Note that now you are working in a python environment, that only knows the modules installed inside the pandas_env environment. Every time you want to use it you have to activate the environment. This might feel a little bit clunky at first, but really shines once you have to manage different versions of python (like 2.7 or 3.4) or you need a specific version of a module (like numpy 1.7).

Edit:

If this still does not work you have several options:

  1. Check if the right pandas module is found:

     `(pandas_env)user@machine:~$ python` Python 2.7.10 |Continuum Analytics, Inc.| (default, Sep 15 2015, 14:50:01) >>> import imp >>> imp.find_module("pandas") (None, '/path/to/miniconda3/envs/foo/lib/python2.7/site-packages/pandas', ('', '', 5)) # See what this returns on your system.
  2. Reinstall pandas in your environment with $ conda install -f pandas . This might help if you files have been corrupted somehow.

  3. Install pandas from a different source (using pip ). To do this, create a new environment like above (make sure to pick a different name to avoid clashes here) but replace point 4 by (pandas_env)user@machine:~$ pip install pandas .
  4. Reinstall anaconda (make sure you pick the right version 32bit / 64bit depending on your OS, this can sometimes lead to problems). It could be possible, that your 'normal' and your anaconda python are clashing. As a last resort you could try to uninstall your 'normal' python before you reinstall anaconda.

If you are facing same problem as mine. Here is the solution which works for me.

  1. Uninstall every python and anaconda.
  2. Download anaconda from here " http://continuum.io/downloads " and only install it (no other python is needed).
  3. Open spyder and import.
  4. If you get any error, type in command prompt

    pip install module_name

I hope it will work for you too

You should first create a new environment in conda. From the terminal, type:

$ conda create --name my_env pandas ipython

Python will be installed automatically as part of this installation. After selecting [y] to confirm, you now need to activate this environment:

$ source activate my_env

On Windows I believe it is just:

$ activate my_env

Now, confirm installed packages:

$ conda list

Finally, start python and run your session.

$ ipython
  1. Another alternative is to use Pycharm IDE. For each project, you can set the Project Interpreter in Settings.

  2. For example, if anaconda is installed in /home/user/anaconda2/bin/python, you can select the Project Interpreter and set to this folder.

  3. Since the whole project is set to Anaconda's path, you can import any module which is packaged within Anaconda.

For OSX:

I had installed this via Anaconda, and had a hell of a time getting it to work. What helped was adding the Anaconda bin AND pkgs folder to my PATH.

Since I use fishshell, I did it in my ~/.config/fish/config.fish file like this:

set -g -x PATH $PATH /Users/cbrevik/anaconda/bin /Users/cbrevik/anaconda/pkgs

If you use fishshell like me, this answer will probably save you some trouble later using pandas as well.

I've had the same exact problem in that I installed Anaconda because a python script I want to use relies on pandas, and that after so doing, python still returned the same comment that "pandas module is missing" or something to that effect.

When I typed "python" to see which python was being called, I found it was still accessing the older version of python 2.7, even though when I installed Anaconda the installer asked (and I agreed) that it would make its python the default python on my machine (PC running Windows 7).

I tried to find if there is a CONFIG.SYS file on the PC, but gave up after searching in various places (If anyone knows, please tell me). I got around the problem by writing a one-line batch script named python2.bat that called the Anaconda2 version of python, which then worked. However, it would clearly be better to change the CONFIG.SYS or whatever the PC uses to decide which version of python to call.

I know there are a lot of answers to this already but I would like to put in my two cents. When creating a virtual environment in anaconda launcher you still need to install the packages you need. This is deceiving because I assumed since I was using anaconda that packages such as pandas, numpy etc would be include. This is not the case. It gives you a fresh environment with none of those packages installed, at least mine did. All my packages installed into the environment with no problem and work correctly.

You can only import a library which has been installed in your environment.

If you have created a new environment, eg to run an older version of Python, maybe you lack 'pandas' package, which is in the 'base' environment of Anaconda by default.

Fix through GUI

To add it to your environment, from the GUI, select your environment, select "All" in the dropdown list, type pandas in the text field, select the pandas package and Apply.

Afterwards, select 'Installed' to verify that the package has been correctly installed.

What worked for me, on my Mac at least, was that I opened PyCharm system preferences, then chose my project on the left side. I clicked on Program Interpreter and looked in the list to see that pandas was not installed. I simply chose it from the list on the right (using the search at the top). I clicked the install package and this resolved the issue.

even after installing anaconda i got the same error and entering python3 showed this:

$ python3
Python 3.6.9 (default, Nov  7 2019, 10:44:02) 
[GCC 8.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.

enter this command: source ~/.bashrc (it is kind of restarting the terminal) after running the command enter python3 again:

$ python3
Python 3.7.4 (default, Aug 13 2019, 20:35:49) 
[GCC 7.3.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> 

this means anaconda is added. now import pandas will work.

pip install module_name 将起作用,或者如果您使用的是以前处理过的文件,而不仅仅是执行 shift+enter 重新加载并完成工作

i had pandas installed ('conda list|grep pandas') and python could find it ('python; import imp; imp.find_module("pandas");'

However still was getting this error inside spyder. I had accidentally been using 'spyder3' instead of 'spyder' command, the former using a different python (3.5) rather than one conda is using (3.8). Running spyder and things worked as expected.

If you installed anaconda as you mentioned then simply type in the anaconda command prompt:

conda install -c anaconda pandas

as mentioned here: https://anaconda.org/anaconda/pandas

Then the import pandas as pd will work.

The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM