23 Years of PyCon
PyCon is an annual Python conference organized for the Python community by the Python community. It is a place of learning, networking and creating. Since the first Python Conference in 1994, PyCon has only grown in size, diversity and breadth of topics. Each year, our team offers the public a PyCon Guide to navigate and save interesting talks. This year, the conference took a shift towards data with talks focused on machine learning, data wrangling, modeling and visualization. Our team was excited to get our hands dirty with data to unlock insights for our clients and expand our capabilities.
It’s a Marathon, Not a Sprint
The conference began with two days of intensive tutorial workshops for both the novice programmer and seasoned professional. The tutorials were followed by 3 conference days jam-packed with keynotes and talks on how people are using Python. Finally, the conference was followed by five days of development sprints where participants can contribute to their favorite open source projects and advance the Python community.
What’s in Your Python Tool-belt?
The latest and greatest tools in the world of Python enable developers, companies and scientists to understand our world and invent the future.
So, what tools can you find in a hacker’s tool-belt, and how is she or he using them to unleash the power of python?
It was recently announced that Python 2.7 will retire after 2020. What if your favorite photo sharing app, and all of your data, went down because of this? There would be no more cute instas of puppies or hilariously relatable memes to double tap! Yes, Instagram, with 400 million active users each day, is built on Python. At PyCon, Instagram shared its realization that if it is to continue growing and evolving, it would have to invest in growing and evolving its tech stack as well. To continue operating at scale, Instagram made the transition to Python3 with zero user experience interruption.
Python is making an impact in the spheres of data analysis and research. Researchers at PyCon showed off their jupyter notebooks to recreate experiments in real-time. These open-source web apps can contain code, equations, explanations and visualizations all in one centralized place. Speaker Katherine Scott demonstrated the power of jupyter with her Python from Space talk, where she aggregated and visualized satellite images of downtown Portland. Why does this rock? Because you can easily reproduce this and other analyses in an environment that is clean, organized and well-tested.
Data Science Stack
Known for its “batteries-included” philosophy, Python allows developers to easily take advantage of the wide-range of robust packages without much legwork. Data scientists are using several statistical and visualization packages to wrangle data into insights with:
GIF by Max Breaux
Cool Datasets and APIs
Some of our favorite talks at PyCon involved exploring interesting datasets with data science and machine learning to draw insights about our world. From AWS public datasets to Planet Labs there is no shortage of cool data to explore. Our team at ISL is particularly excited to explore and visualize satellite imagery here in the District!
Mo Data, Mo Problems
All of this data talk inspired our team to create our latest open-source project mo-data. Mo-data is a cookiecutter template for ISL data projects that sets up a standard data science stack featuring Anaconda, Jupyter, NumPy, SciPy and Bokeh. Be on the lookout for our first release on our GitHub. We’re excited to wrangle some data, are you?
Were you at #PyCon2017? What’s in your Python tool-belt? I’d love to hear how you’re changing the world through data. Tweet at me @juliebaconator or email us at email@example.com.