Unmesh Mali

Week 20, 2021

The highlight this week was definitely my bike commute while returning from work. It’s a 7.5 miles commute with challenging elevation gain. I am used to it except SF experienced unusually high winds and I tried to go against it. But gave up and put my bike on a bus and returned home.


Work

Work was slow this week. I worked on and finished only one analytical project. Rest of the tasks were reading, working with others with troubleshooting, and brainstorming a control logic to minimize green house gas emissions for a group of houses. I wish I was doing more analytically technical tasks but it is what it is.

Projects

Machine Learning Bootcamp

This bootcamp got very busy super fast. Currently studying linear regression with all its nuances. Uni-variate regression, multi-variate regression, using gradient descent to converge a model. Also completed 3 projects in the last week. Covid-19 infections predictions for Chicago, Covid-19 international cases and tests analysis and prediction, housing prices prediction using gradient descent. Here’s the link to my Github profile where I have uploaded the projects. https://github.com/unmeshmali25

CEEW – Sustainable Mobility Research Assistant

I started helping out CEEW with a hackathon type competition. The project is to make rapid assessment of bus stops in terms of accessibility and safety. So, looking at things like private cars parked in front of the bus stop, benches to sit on, curbs around the bus stop, lights etc. The project is on a very tight schedule. We have just over a month to deliver on this project. I started working on collecting data. Chasing 2 options right now. - Get locations of all the bus stops in an Indian city and programmatically download Google Street View Images at those locations. - Use web scrapping to download images that show up in google image search after typing in “bus stops in India” keyword. I have executed on the latter option. Next week, we’ll have to focus on using pre-trained models to run on our test images. Given that we don’t have a lot of time, going with pre-trained neural network models is out best bet.

Books

Couldn’t read anything this week 🙁

Fitness