Hackathon

Analytics Vidhya Workshop / Hackathon – Experiments with Data

Anirudh Hackathons , , , , , ,

This was a hackathon + workshop conducted by Analytics Vidhya in which I took part and made it to the #1 on the leaderboard. The data set was straight-forward and quite clean with only a minor need for missing value treatment. This post will might be useful for people who want a walk-through on the steps involving data munging and developing machine-learned models.

screenshot-datahack.analyticsvidhya.com 2016-09-01 23-43-54

 

The workshop ended with a basic hackathon with data given on age, education, working class, occupation, marital status and gender of individuals and one had to predict the income bracket of these individuals.

I’ve posted the data and my code and solutions in this GitHub repo. An IPython Notebook has also been shared.

I approached the problem first by attempting some feature engineering (other than missing value treatment) on the data, and then ran a basic logistic classifier and a random forest classifier. However it turned out that these models performed better without feature engineering, which shows the dataset was already quite clean and informative to begin with for this competition.

I later attempted gradient boosting with parameter tuning to maximizing scores.

Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
view raw experimentsWithData.ipynb hosted with ❤ by GitHub
Advertisement

MITx 15.071x (Analytics Edge) – 2016

Anirudh Non Technical , , , , , ,

I am auditing this course currently and just completed its 2nd assignment. It’s probably one of the best courses out there to learn R in a way that you go beyond the syntax with an objective in mind – to do analytics and run machine learning algorithms to derive insight from data. This course is different from machine learning courses by say, Andrew Ng in that this course won’t focus on coding the algorithm and rather would emphasize on diving right into the implementation of those algorithms using libraries that the R programming language already equips us with.

Take a look at the course logistics. And hey, they’ve got a Kaggle competition!

AnalyticsEdgeLogistics

There’s still time to enroll and grab a certificate (or simply audit). The course is offered once a year. I met a bunch of people who did well at a data hackathon I had gone to recently, who had learned the ropes in data science thanks to Analytics Edge.

My First Data Science Hackathon

Anirudh Non Technical , , , ,

So after 8 months of playing around with R and Python and blog post after blog post, I found myself finally hacking away at a problem set from the 17th storey of the Hindustan Times building at Connaught Place. I had entered my first ever data science hackathon conducted by Analytics Vidhya, a pioneer in analytics learning in India. Pizzas and Pepsi were on the house. Like any predictive analysis hackathon, this one accepted unlimited entries till submission time. It was from 2pm to 4:30pm today –  2.5 hours, of which I ended up wasting 1.5 hours trying to make my first submission which encountered submission error after submission error until the problem was fixed finally post lunch. I had 1 hour to try my best. It wasn’t the best performance, but I thought of blogging this experience anyway, as a reminder of the work that awaits me. I want to be the one winning prize money at the end of the day.

🙂

screenshot-datahack analyticsvidhya com 2015-12-20 18-41-12

 

Which Programming Languages Get Used Most At Hackathons?

Anirudh Non Technical , ,

For a quick peek into the list:

The Top 10 Languages At Devpost’s Hackathons:

  1. HTML/CSS (see note below)
  2. JavaScript
  3. Python
  4. Java
  5. C/C++
  6. PHP
  7. Objective-C
  8. C#
  9. Swift
  10. JSON (which isn’t … really a programming language, but is on their list for some reason, so I’m including #11 too)
  11. Ruby

Read the full Techcrunch article to know why.

In stark contrast:

The Top 10 Languages according to IEEE Spectrum’s 2015 Rankings:

  1. Java
  2. C
  3. C++
  4. Python
  5. C#
  6. R
  7. PHP
  8. JavaScript
  9. Ruby
  10. Matlab

Note: HTML isn’t quite a “programming” language — it’s a markup language, meaning it’s a means of laying out the elements of a document. But it’s a “language” none the less, and one that pretty much every web developer taps endlessly, so we’ll let the semantic stuff slide