The best plotting library I've found for Python. Has a steep learning curve if you need precise control, but can do just about anything. I've published charts in scientific journals using matplotlib, and they looked great.
This is a wonderful set of tools. If you aldeady know python and Matlab learning this is a breaze, the syntax is very similar. The resulting plots are publication quality, suport of LaTeX fonts is very good, and the plots are easy to script so that making small changes months after you produce the original is easy. I have never had any issues with the plots being inaccurate.
Well I spent a couple hours trying to figure out why the plotted value and the calculated value were so different. I kept changing the resolution, but no dice. Finally, I copied the code into MATLAB, with a few syntax changes, and ran it. Boom, right answer. Sure it's a little harder to learn than MATLAB, but I figured it would be worth it. It doesn't matter what it costs or how powerful it is, in scientific computing if it's not accurate it's useless. Very disappointed. Especially considering the effort I'd already put into learning it. In my opinion, scientific computing in Python is still a bit mickey mouse. But if your time's not worth anything to you, and you want a skill that no employer's looking for, go for it.
Powerful plotting tool and not really hard to learn. Combine it with ipython. It becomes really powerful if you know a bit of python, but you can start even without.
Steep learning curve, but a significant lot of benefits awaits you at the far end.
The syntax may look a bit difficult in the very beginning. But once you get used to it, you can plot your data with infinite possibilities. I use matplotlib in my daily scientific work and it rocks!!!!