I'm trying to install scipy and numpy on Ubuntu 16.04 but I keep getting the following error. Can anyone tell me how to install the dependencies?

Reading package lists... Done
Building dependency tree       
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 python-numpy : Depends: python:any (>= 2.7.5-5~)
 E: Unable to correct problems, you have held broken packages.
  • Make sure your repo sources are up to date. Run the following command to fix broken or missing dependencies. sudo apt install -f Above command will only download the missing dependencies if you have already installed the package. – sgiri Jan 6 '17 at 8:11

You can also use pip (the alternative Python package installer) to install numpy and scipy for the whole system:

sudo apt-get install python-pip  
sudo pip install numpy scipy

This could install it regardless of dependency errors in the Ubuntu package manager.

  • 1
    Would this conflict with any updates if you had already installed numpy/scipy earlier and tried using system-wide pip? – Ruslan Jun 9 '19 at 9:13
  • If you installed them earlier with pip, then this command will not reinstall them unless you explicitly tell it to upgrade the packages. – don.joey Jun 10 '19 at 6:42
  • 1
    I mean if you had this package installed previously. – Ruslan Jun 10 '19 at 7:58
  • Perhaps this answer is for python2 because it didn't work for python3. karel's answer worked for python3. – Ramid Jan 6 at 11:39

To install the dependencies in all currently supported versions of Ubuntu open the terminal and type the following commands:

sudo apt update  
sudo apt install --no-install-recommends python2.7-minimal python2.7  
sudo apt install python-numpy # in Ubuntu 20.04 and earlier
sudo apt install python-scipy # in Ubuntu 18.04 and earlier

For Python 3.x

sudo apt update  
sudo apt install --no-install-recommends python3-minimal python3  
sudo apt install python3-numpy python3-scipy
  • I'm still getting the same error. – Rahul Jan 6 '17 at 10:18
  • python --version gives Python 2.7.12 – Rahul Jan 6 '17 at 10:20
  • python-numpy : Depends: python:any (>= 2.7.5-5~)is the error message in your question. You have a more recent version of Python 2.7.12 installed, however you are still getting the same error anyway. – karel Jan 6 '17 at 10:28
  • This is the real answer. Always use apt (the same as apt-get) if you can, it will not lead to conflicts, while pip will sometimes disturb your dependencies and versions. – questionto42 Mar 18 at 21:48

In my case I wanted scipy installed into a virtual environment instead of globally. Installing libatlas-base-dev and gfortran prior to pip install resolved the issue:

sudo apt-get install libatlas-base-dev
sudo apt-get install gfortran
source .venv/bin/activate
pip install scipy

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.