Datmo: A CLI to harness power of tracking ML models
Piyush Gupta (~piyush2) |
Delving into the vast, ever-growing realm of data science may seem daunting at first, to beginners who are looking to delve into it. There are so many online courses to take, libraries to learn but no fixed approach to doing so. Why? We asked ourselves this too. The two major problems are accessibility and awareness.
That’s what Data Science for India (DSI) is seeking to change. Founded in UC Berkeley, the organization strives to make data science (and data literacy in government schools) more accessible in India. This summer, our volunteer instructors taught our introductory data science course - which goes from basic statistics and programming constructs all the way to elementary machine learning algorithms.
When DSI was just getting started, we got in touch with a Berkeley graduate who shares our vision of democratizing access to machine learning through building Datmo – a CLI (like Git), that allows users to track machine learning models and run them efficiently.
In this talk, we will be walking you through a jupyter notebook that a goes into more detail on a question taken straight out of DSI's introductory data science curriculum, in which we will use linear regression to predict the salaries of engineers given their exam scores - and using Datmo, keep track of our models with a just a couple lines on our terminal.
We hope that you will enjoy deep diving into machine learning and, running, tracking, and sharing your work with Datmo. Together, let's carry forward Data Science for India's initiative to spread knowledge of this powerful and growing field to the youth of our country, and take our country's future to the next level.
Python 2 and 3, Github account, macOS/Ubuntu system (Not compatible with Windows yet)
Github repo: https://github.com/dsindia/datmo-workshop
Piyush Gupta (Undergraduate at IIT Delhi)
I am an aspiring data scientist pursuing an undergraduate degree from IIT Delhi (currently in junior year) and have experience working at an IoT startup and a telecom major (TCL) with implementing machine learning and data science pipelines. This summer, I joined Data Science for India (DSI) which aims to spread data literacy in India by getting high schoolers interested in the vast cosmos of data science methods though accessible lessons given by our volunteers across India!