Insights into our prediction models

Jul 11, 2020

We believe in socially responsible research. We are answering some questions related to our work so that the technical community can understand what we are doing and come up with ways to help.

Introducing the first Covid19 future tracker for India

Jun 25, 2020

Forecasting is a difficult task. Take weather for example. We have seen how accurate the weather department is with rainfall predictions. The patterns return year after year, all the patterns have been studied and models have been made. They have access to satellites and supercomputers. Still they get it wrong so many times.

Can you guess which state will end up having the most Covid-19 cases?

Jun 11, 2020

Currently we are seeing that Mumbai, Delhi, Chennai and Ahmedabad together account for 2/3rd of the cases. But the same trend would not continue for long. When all of this is over, UP would end up being the state that would report the maximum number of daily cases.

Training a 20 day forecaster

Jun 04, 2020

The 10 day forecaster that we talked about in the previous post, allows us to predict only the next 10 days of Covid19 new cases by looking at the past 20 day trend. Greater the number of days that we can predict, the more is the time that we get to prepare for the future. Hence we enhanced our neural net to learn from 40 day trends and predict next 20 days.

10 day forecast using Deep Learning

Jun 04, 2020

Most researchers are using a compartmentalised model such as SIR or one of its variants to predict the Covid19 future, because the epidemiologists use them to describe epidemic growth. These work well when there is unhindered growth of the disease and when there is a uniform policy and pattern in the population under study. We felt that it was too much of a simplification to a heterogenous population like India where the policy and implementations are changing across states and with time.

The race to the top

May 17, 2020

A pandemic is just a game of numbers. The Covid19 virus does not discriminate based on your nationality, religion or political affiliation. China is the most populous nation but they seem to have succeeded in containing the disease. India hasn’t. So it is just a matter of time till we reach the top.

What has the lockdown achieved? - Part I

May 12, 2020

With Lockdown 4.0 looming over, in most of the states, you may be thinking what is it that we have achieved.

Prediction of India's Covid19 growth using a Neural Network (AI)

May 11, 2020

On April 26th, we analysed the number of new cases per day and built a model to estimate the rate at which Covid19 cases will grow in India.

Request for data for Medical Resources dashboard

May 10, 2020

CovidIndia open source tracker has become quite popular. We are a dev team of volunteers who have added to the code base and have this staging server where we track 5 new medical resources - the number of doctors, nurses, ICU beds, ventilators and total beds at different granularity.

Verification of Peak Death dates by University of Washington

May 08, 2020

A peak date is when the epidemic growth is at the maximum and after that the disease starts to subside. Hence it is of interest to know if a country or state has peaked.

Has any state in India won over Corona yet?

May 07, 2020

This is a question that every curious Indian wants to know. Apart from the political claims which we consider to be just noise, is there a scientific way to actually confirm this?

Has India Peaked Yet?

Apr 30, 2020

On 26th April, Data Driven Innovation lab (DDI) at Singapore University of Technology and Design (SUTD) came out with an implementation of the popular SIR model and tried to predict the Covid19 peak dates for various countries including India. They predicted that India had already peaked on 20th April and 97% of the expected cases would have been recovered by May 25th. We were curious to verify these dates and built a methodology to validate these predictions.