Hacker News Reader

Comment by plafl | Hacker News Reader

It's a good library, but I have the feeling that it tries to do too many things, or that you can do the same thing in too many different ways.

View on HN

Build & installation from scratch is far too complex. Relies on fortran libraries of numpy which I can't get built inside my corporate domain

Why is that panda's problem? If you have scipy and numpy installed, it's just sudo pip install pandas....

You can't do that in a closed envrionment

> You can't do that in a closed envrionment

This is the real killer. Any leads on making it less closed?

Also, I'm unfamiliar with how pip works, but you can't even install into your local user profile (i.e. w/o root)?

Sometimes, even when installing to a local virtual environment, if there are required libraries that require, for example a Fortran compiler, you'll need to install that on the system.

Use Anaconda Python distribution (https://store.continuum.io/cshop/anaconda/). Comes bundled with pandas, numpy, scipy (and much more). Does not require admin privileges. Updates and installations do not require working compiler (binary packages for your platform are downloaded).

Thanks. Tried this and couldn't get it to run on RH5.5

What issue did you have? Please email info@continuum.io with the information - we would love to know more about your problem and help you resolve it.

Just FYI, when 'sudo pip install pandas' doesn't work (which it didn't for me recently), you'll get no love upstream:

https://github.com/pydata/pandas/issues/7517

As @mynegation notes, you can use Anaconda (or virtualenv).

Reply on news.ycombinator.com