numpy - Build a stable numerical computing environment with Python on Linux -



numpy - Build a stable numerical computing environment with Python on Linux -

i working on phd thesis , should finish results section next week. since did apt-get upgrade, matplotlib not work anymore:

importerror: cannot import name '_tkagg'

i found several solutions. 1 of them install python-tk , rebuild matplotlib source, failed due reasons not know more (have seek again).

is there way build stable python, numpy, scipy, matplotlib1, pandas environment on linux (debian). never updated. or somehow "stable" numerical computing environment.

edit:

what using virtualenv in order build independent , stable/working python environment ?

i used numpy/scipy under debian thesis too.

if want stable software, find works , don't upgrade it. otherwise... bugs happen, if you're running debian unstable.

i realize many of run apt-get dist-upgrade willy-nilly because latest features in every package, there's no free lunch: if want latest features, latest bugs, incomplete upgrade scripts, etc.

python numpy matplotlib scipy

Comments

Popular posts from this blog

php - Android app custom user registration and login with cookie using facebook sdk -

django - Access session in user model .save() -

php - .htaccess Multiple Rewrite Rules / Prioritizing -