# Installing and updating packages Anaconda comes with a lot of scientific packages, however if you need an additional package or want to update one, you can use `conda` or `pip`. ## Update Anaconda Distribution Anaconda publishes updates to its whole distribution every few months. Start _Anaconda Prompt_ and update then by ``` conda update conda conda update anaconda ``` ## Installl new packages Many packages are hosted on [Anaconda Cloud](https://anaconda.org/), you can search there for a package and follow the install instructions. Typically you can install a new package with the following command inside _Anaconda Prompt_: ``` conda install -c conda-forge tensorflow ``` or ``` pip install tensorflow ``` ## Conda vs pip pip: - default python installer - has always up to date packages - usually maintained from the package owners - recomented for pure python packages conda: - general purposed package manager - mainly used for python - some packages are outdated - recommented for libraries that have C++ dependencies (`numpy`, `scipy`, `matplotlib`, `cython`, `numba`, ...) - These are well maintained from anaconda - numpy is usually faster when installed from conda