Have you ever sent a Python script to someone, only to have them complain about missing libraries, or incompatible versions? Have you found yourself writing long-winded HOWTO docs around how to use your scripts? We’ve seen many scientists and analysts have this problem, and they are real impediments to robust, repeatable collaboration.
By using Wakari’s Bundles, these problems go away entirely.
A bundle is a collection of files shared as a directory. Bundles are centered around the IPython Notebook, which nicely allow us to mix narrative documentation alongside actual code and output. In addition to the Notebook file itself, bundles can also include Python modules and data files. Bundles shared on Wakari have access to all of the packages installed in Anaconda. Since relative imports work inside of bundles, you can build sets of modules without going through the complication of creating a full Python package and writing a setup.py file.
There are myriad possibilities for using bundles. In finance, they can be used to share code and analysis in lieu of emailing Excel spreadsheets. For Customer Acquisition Analytics, you can easily integrate many data sources (Google Analytics, Databases, Mixpanel), because at the core of the bundle you have executable Python code. In academia research groups, bundles can be used to share analysis and data for reproducible science. We have already seen professors using bundles for classes.
It’s easy to make a bundle, and in this blog post we will walk you through the process.
Click the share button next to the IPython notebook or directory that you would like to share.
At this point you will be asked to name the bundle:
Select which IPython notebooks will be displayed on the view page. Here you can also password protect your bundle if you have a paid account.
Click the submit button.
You should immediately see a sharing process dialog. Wait for the sharing process to finish. If you encounter an error, please contact firstname.lastname@example.org.
Finally, when sharing process is complete, you will be given a link to the bundle. You can see all of your bundles at https://wakari.io/[username], mine are at https://wakari.io/paddy
Running code from a bundle is easy too. Just go to a bundle URL like Lecture 2 NumPy and click “Run/Edit this in Wakari”. This will proceed to copy this code into your Wakari account and open the main IPython notebook of the bundle. If you haven’t created an account yet, you will be prompted to create one in the signup process. Free accounts can create public bundles; paid accounts can create public and private bundles.
We are really excited about bundles on Wakari, and we can’t wait to see what our users will create. Sign up at https://wakari.ioTags: Wakari IPython Notebook Packaging comments powered by Disqus
Back to Blog→
- Anaconda Server
- Bayesian Data Analysis
- Big Data
- Boolean satisfiability problem
- Data Science
- IPython Notebook
- Open Source
- Practical Python
- Product Release
- Product Update
- Python 3
- Raspberry Pi
- SAT solver
- Social Media
- Thomson Reuters
- White Paper