For more information about using the conda package manager in Anaconda Prompt (Terminal on Linux or macOS), see the conda documentation. If conda cannot find the file, try using an absolute path name instead of a relative path name. Problem With the release of Anaconda 5.0, a new channel called main has been added to defaults. The new channel is given the top priority within defaults and holds packages built with the new compiler stack. Conda versions prior to 4.3.26 do not have the main channel included as part of the defaults channels.
To do my web crawling, I started using,which is a Python module for doing web crawling. I installed it from my command prompt by doing ‘pip install selenium’, and Selenium was working just fine in PyCharm and Python shell. But when I tried doing ‘import selenium’ in a Jupyter notebook, I kept getting a module not found error. It turned out to be a Python path issue. In short, Selenium had already been installed, but Jupyter could not import Selenium because it wasn’t pointing to the path where Selenium had been installed.
The usual digging led me to, but in the end I couldn’t get Selenium to work on my Jupyter notebook by following the instructions provided. Fortunately, an intern told me about, a GUI-based application that could be used to install packages for the (virtual) environment running Jupyter notebook. As long as you have anaconda installed, you just had to run the below command to install Navigator: So I tried searching for the Selenium package on Anaconda Navigator, but the search returned no results. After doing some digging, I came across that had a piece of code I could run to get Selenium: conda install -c conda-forge selenium After running this code in my command prompt, I got selenium to work on Jupyter notebook! Now I could use Selenium and Chrome Driver in a Jupyter notebook just fine.
News about the dynamic, interpreted, interactive, object-oriented, extensible programming language Python If you are about to ask a 'how do I do this in python' question, please try, or the #python IRC channel on FreeNode. Please don't use URL shorteners. Reddit filters them out, so your post or comment will be lost. User level install of the version of python you want. Able to install/update packages completely independent of system libraries or admin privileges. conda tool installs binary packages, rather than requiring compile resources like pip - again, handy if you have limited privileges for installing necessary libraries. More or less eliminates the headaches of trying to figure out which version/release of package X is compatible with which version/release of package Y, both of which are required for the install of package Z.
Comes either in full-meal-deal version, with numpy, scipy, PyQt, spyder IDE, etc. Or in minimal / alacarte version (miniconda) where you can install what you want, when you need it.
No risk of messing up required system libraries. User level install of the version of python you want Pyenv. Able to install/update packages completely independent of system libraries or admin privileges pip install -user or virtualenv.
conda tool installs binary packages, rather than requiring compile resources like pip - again, handy if you have limited privileges for installing necessary libraries. Wheels are making this less and less of an issue. More or less eliminates the headaches of trying to figure out which version/release of package X is compatible with which version/release of package Y, both of which are required for the install of package Z This isn't an issue with pip either.
Comes either in full-meal-deal version, with numpy, scipy, PyQt, spyder IDE, etc. Or in minimal / alacarte version (miniconda) where you can install what you want, when you need it OK. No risk of messing up required system libraries Unless you're working with admin or root privileges I don't see how you'd manage that. Suffice to say that I'm not really seeing the huge draw. Suffice to say that I'm not really seeing the huge draw. The same reason that people comfortable writing code in Eclipse or emacs don't see the draw of Pycharm or visual studio. If Anaconda doesn't provide a benefit, no problem, there's no obligation to use it or even try it.
Requests doesn't provide anything not already available in the standard modules, but that doesn't mean that people don't appreciate the effort in providing a simpler packaged solution. Same with Anaconda vs installing everything from pip. Does anaconda have a dramatically better user experience than pip and virtualenv?
Is anaconda an improvement for people already experienced using different tools in the python ecosystem that solve the same problems? The answers are yes and yes, if they are qualified with 'windows doing science'. At least a few years ago, getting both matplotlib and numpy up and running was nontrivial, and ran the risk of discovering dll hell. Also, irrespective of OS, if there is a need to maintain legacy code in a variety of minor versions of the interpreter (e.g. 2.5, 2.6.2, 2.7.10) (looking at you, science!), this is rendered trivial, since the interpreter is treated as simply another package. If you're developing in a Windows environment the most significant advantage is ease of installation. Although, it's a lot easier for someone new to python to install one Anaconda package, even if it's 350MB, than to hunt down each individual package and install the wheel with pip.
This has the advantage of lowering the pain threshold for someone who wants to experiment with ipython notebook or pandas or some other useful tool and will likely result in them starting down the road to learning Python. If they can't even get numpy installed because of pip compiler error, they're more likely to not experience that initial thrill and not pursue the language. I've been enjoying working on OS X. Office applications with good enough compatibility with the Windows versions for what I do (although apparently the next Mac release is going to finally bridge that gap) so that I'm able to work in that environment when necessary, but have the benefits of running on.nix the rest of the time. Most packages that work on other.nix platforms should either be compatible or be easy to find someone detailing the minor tweaks you need to get it working; I think only once did I find myself stymied trying to get a Unix package working on OS X. (I do keep Windows installed in Parallels for times where there just isn't a Mac option, but I prefer to not have to use it when I don't have to.).