OSCON 2014
If you haven't installed Anaconda, go here: https://store.continuum.io/cshop/anaconda/
If you have installed a distribution like Anaconda or Enthought, you have IPython notebook. To run IPython notebook and to follow along:
- git clone this repo OR download it as a zip and unzip it to your Desktop
- open the terminal/command line and cd into hands-on-analysis-python/notebooks
- in your terminal, type: ipython notebook --pylab
If you didn't install these tools with a distribution, you can still run the code files in the code/directory in your terminal. You'll have to uncomment the print statements, and recomment if you wish, as we go.
To run, cd into hands-on-analysis-python/code and type python pandas_lessons.py
Wakari.io is a platform for web-based Python data analysis. It also allows you to host IPython Notebooks online and share them so people can run IPython Notebook in the browser. To sign up for Wakari:
- go here https://www.wakari.io/
- when you register, it might give you a message that "we're at capacity for this plan" or something. Ignore it and click on the Wakari logo in the top left corner.
- Then, click this link: https://www.wakari.io/sharing/bundle/sarguido89/hands_on_data_analysis
- Click on "run/edit this notebook"
- It's going to say "404: Not Found". That's okay! On the left, double click on "notebooks", and then on the notebook you want to open.
- You should be good to go
If none of these options work, or you just want to watch now and code later, you can follow along below. It's not interactive, but you'll be able to see what's happening.
- Pandas: http://nbviewer.ipython.org/gist/sarguido/b91f917a8cd443eb33ad
- Scikit-learn: http://nbviewer.ipython.org/gist/sarguido/8191b91a79fbc41c439b
- Matplotlib: http://nbviewer.ipython.org/gist/sarguido/b2862f4d5f5da5655c84
https://docs.google.com/document/d/1tY6ifr4m4h2Ri445Vro1ZKr7Kap0X1IcI2OGl7IDaU0/edit?usp=sharing
Q: HEY, why didn't you cover THAT THING?!
A: Because I didn't, okay?! I only had three hours! Three hours is not enough. You could do a six hour workshop on each tool and still not cover everything.
Q: Your visualizations are NOT great.
A: Yes, I know...but now you know where to start to make them better!
Q: But really, why didn't you cover unsupervised learning/getting rid of missing values/making time series plots/etc etc?
A: Aw man! There are just so many things!
Q: What's your twitter handle again?
A: @sarah_guido