Why isn't Python built in advance with the required libraries like pandas, numpy, etc.
What is the reason for distributing packages separately?
- Why do we have separate "extra" packages like pandas, numpy?
- Since these modules seem so important, why aren't they part of Python itself?
Are Python "single distributions" available for preloading?
-
If this is part of the design, in order to keep the "core" separate from the extra functionality, it should still be at least "pre-imported" as soon as you start Python.
-
Where can I find such distributions, if they exist?
source to share
Many of these tools, including core Python, are developed and distributed separately by different teams, so aggregators can curate them and put them on the same distribution. Here are some notable examples:
- Anaconda distribution from Continuum Analytics
- Canopy distribution from Enthought
- ActivePython from ActiveState
- PythonXY for scientific programming
- WinPython for Windows
- PyIMSL Studio by Rogue Wave Software
source to share
You can do an interactive interpreted launch with "pre-imported" modules as well as pre-launch code using the Interactive start file .
Alternatively, you can use the Configuration Modules to pre-run your code on every python call.
As to whether pandas and numpy should be part of the standard library is a matter of opinion.
source to share
It's a bit like asking, "Why isn't every motor equipped with a car?"
While a car without an engine is pretty useless, the inversion is not performed: most engines are not even used for cars. Of course, you can try to sell a full car to people who want a generator, but they won't buy it.
Also, people who design cars may not be the best at building an engine and vice versa.
Similar to python. Most python distributions are not used with numpy, scipy or pandas. Distributing python with these packages would create a huge overhead.
However, there is of course a high demand for pre-built distributions that combine these modules with the appropriate python and ensure that everything will interoperate smoothly. Some examples: Anaconda, Canopy, python (x, y), winpython, etc. Thus, the end user who just wants the car to work is best off choosing one of them rather than having to install everything from scratch. Other users who want to always have the latest version can choose themselves.
source to share