Update as of 21:04

Fixed the SEIR Model

I fixed the SEIR model! The US has passed its inflection point as of 4/10 or two days ago. THIS IS WONDERFUL NEWS!!!!

Here, look at it yourself, and remember the plots are interactive. Go down to the curve for the US: SEIR Model Updated

Here’s the global curve:

And NY:

Update on Sigmoid Fit Model

Updated today with data from 4/11/2020

Ok, so I’ve been trying to figure out how to embed the plotly plots from my code onto this webpage but it’s impossible. I want you all to see the plotly plots because they are interactive and some are time-lapse visualizations of corona taking over the world. .png files just do not do it justice.

The last time I posted my EDA and Sigmoid fit code, there was simply too much code to get through if you wanted to see the plotly plots. You’d literally have to hit “end” on your keyboard and go backwards. So I first re-wrote the code and shifted the cells with the plots up top. Upon exporting the code as html, the plots disappeared. So I re-wrote it again out of pure rage. This time I tried to keep the verbosity low.

But please, check out this code and interact with the plots and models I’ve made it will really elucidate your understanding of this pandemic.

Here it is, one more time: Updated today with data from 4/11/2020

About my SEIR model post, the important plots didn’t come up for that .html file either. What’s worse is that my model for the US is very wrong. I got like 250 mil total infected in the US at the end of it all but only 40 mil globally? Definitely messed something up there.

I also have that RNN model which I didn’t get to upload/finish yesterday, as well as an XGBoost model that I have to re-write (because of aforementioned issue), and then I’ll fix the SEIR model because I am positive that will be the most accurate.

It’s just that as of now, the R_0 is like 6 for the US (compared to like 3.8 in Italy, and nearly 8 in NY) so that gives it such an exponential boost that it’s all my model has to work on.

Background

“In response to the COVID-19 pandemic, the White House and a coalition of leading research groups have prepared the COVID-19 Open Research Dataset (CORD-19). CORD-19 is a resource of over 45,000 scholarly articles, including over 33,000 with full text, about COVID-19, SARS-CoV-2, and related coronaviruses. This freely available dataset is provided to the global research community to apply recent advances in natural language processing and other AI techniques to generate new insights in support of the ongoing fight against this infectious disease. There is a growing urgency for these approaches because of the rapid acceleration in new coronavirus literature, making it difficult for the medical research community to keep up.”

The data is available here: