One of the easiest ways to install and work with Python is through Anaconda, and this is the option we will be using for the workshop. Anaconda is an open source Python and R distribution which includes a hundred of the most popular Python, R and Scala packages for data science, as well as conda, a package dependency and environment manager that gives you access to over 720 ready-to-install packages. Anaconda is available for Windows, OS X and Linux.
- If you are comfortable working with Python already and prefer using your own installation instead of using Anaconda, you are free to do it, but we will follow the class assuming you know how to create your own virtual environments and install libraries. Use Python 2.7 to create your environment and install the libraries specified in the paragraph 3.2.
- If you already have Anaconda installed, you can jump to paragraph 2, but remember updating your conda distribution with the following commands:
conda update conda
conda update anaconda
To install Anaconda, you can choose between these two ways, depending on the available space in your computer and the WiFi conditions:
This installs the whole distribution with the most popular data science packages included. Requires 3 GB of available disk space.
Use the following link to download the package according to your operative system and architecture (in any case select the Python 2.7 version): http://continuum.io/downloads
Follow the default installation. If you are installing through your command line and you are asked the following question, we recommend you to answer yes:
Do you wish the installer to prepend the Miniconda2 install location
to PATH in your /home/malonfe/.bashrc|.bash_profile ? [yes|no]
It is a lighter alternative to Anaconda. Miniconda comes just with Python and conda manager. It is an ideal option for users who want to start with a minimal installation and install the packages as they are necessary. Requires 400 MB of available disk space.
Use the following link to download the package according to your operative system and architecture (in any case select the Python 2.7): http://conda.pydata.org/miniconda.html
Follow the default installation for your operative system here. If you are installing through your command line and you are asked the following question, we recommend you to answer yes:
Do you wish the installer to prepend the Miniconda2 install location
to PATH in your /home/malonfe/.bashrc|.bash_profile ? [yes|no]
Using virtual environments when working with Python is a good practice as it means working like in a separate "box" where you can install the libraries you need for a project, not interfering with other environments or projects.
In your terminal write just the command:
conda
If you received a message like this:
-bash: conda: command not found
then keep reading how to add conda to your path. If otherwise the command responses with a lot of options, go to step 3.2.
You need to add the following lines to your /Users/<my-user>/.bash_profile file (changing <my-user> for your user name and <anaconda-folder> for either miniconda2 or anaconda2):
# added by Miniconda2 4.1.11 installer
export PATH="/Users/<my-user>/<anaconda-folder>/bin:$PATH"
Then write the following command for the chages to take effect:
source .bash_profile
If the conda command keeps returning the same error close the terminal and open it again.
You need to add the following lines to your /Users/<my-user>/.bashrc file (changing <my-user> for your user name and <anaconda-folder> for either miniconda2 or anaconda2):
# added by Miniconda2 4.1.11 installer
export PATH="/home/<my-user>/<anaconda-folder>/bin:$PATH"
Then write the following command for the changes to take effect:
source .bashrc
If the conda command keeps returning the same error close the terminal and open it again.
You need to add the route to your anaconda or miniconda bin folder to your PATH in the environment variables. Check this link on how to manage your environment variables in different Windows versions.
To create your virtual environment with conda, run the following command in your working folder:
conda create --name myvenv pandas jupyter matplotlib seaborn
The name of your new virtual environment will be myvenv and you will be adding to it the following libraries: pandas, jupyter, matplotlib and seaborn. This process will take longer with Miniconda as it needs to download the libraries before adding them to the environment.
In case you are using your own Python installation, the libraries you should install in your virtual environment are the same. Some of them depend on others, so installing
seabornwill probably installpandasandmatplotlibalready for you. Remember usingpip freezeto check what libraries you have installed.
Now you need to activate the virtual environment you just created. The command that will do that for you is:
source activate myvenv # Linux and Mac OSX
activate myvenv # Windows
If the virtual environment was successfully activated you should see that the prompt in your terminal has changed to something like this:
(myvenv)<your-computer-name-and-user>$
Now you are inside your virtual environment and you are ready to start working on it.
If you use your own python remember to activate your environment with
source myvenv/bin/activate
All the files that you need are in PyLadiesDC GitHub online repository and can be donwloaded easily:
- Go to the workshop repo https://github.com/PyLadiesDC/python-for-journalists
- In the 'Clone or Download it' button select your choice:
- If you don't use git regularly just use the option 'Download Zip' and unzip the files in your computer
- If you are familiar with git, just clone the repo if you want
The Jupyter Notebook is a web application that allows you to create and share documents that contain live code, equations, visualizations and explanatory text. We will use it to run our Python scripts and check the output of our code. We'll provide you the script you'll need in order to follow the workshop.
With your virtual environment activated, go to the folder you just downloaded and run the following command:
jupyter notebook
A server will be launched and a new tab will be open in your default browser (by default in http://localhost:8888/) allowing you to navigate your scripts, open them, work with them, and save all your changes.
The server takes the control of your terminal, so if you need to run any command on it, either you stop the server with CTRL+C, or you send the process to the background (checkout this link). If you are not really familiarized with unix commands we recommend you to stop and relaunch the server any time you need it, or open two terminals. In the latter case do not forget to activate your virtual environment in both of them.
Check again the conda cheat sheet we provided to learn how to manage the environments. In order to deactivate your virtual environment (you can always simply close the terminal) you should run the following command:
source deactivate
For standard python environments with the
deactivatecommand is enough.
If you ever need to use your system's Python, instead of the Anaconda's Python we have been using today. Remember that we updated your PATH, so anytime you run the python command you'll be using the Anaconda one. In order to avoid that, you can comment the lines of code we have added to either your .bashrc or your .bash_profile file. Or, for Windows systems, delete the route you just added to your path.
If you want to run condas without having the path updated you may:
- either use the absolute path to the command, see the example:
<miniconda2|anaconda2>/bin/conda create --name newenv pandas
- update the path temporarily (for your console session) in the console you are using writing the same line you had in your bash file:
export PATH="/Users/<my-user>/<anaconda-folder>/bin:$PATH"
An useful command to know which Python you are running at any time is the following one, that tells you the path to the executable Python you are calling.
which Python
When using Anaconda the result would be something like this:
/Users/<my-user>/<miniconda2|anaconda2>/envs/myvenv/bin/python