search

CodeForPhilly / chime

COVID-19 Hospital Impact Model for Epidemics
https://codeforphilly.github.io/chime/
MIT License
205 stars 151 forks source link

app.py #455

Closed stanbasickes closed 4 years ago

stanbasickes commented 4 years ago

Summary

Additional details

Suggested fix

PhilMiller commented 4 years ago

Please don't open insubstantial issues on the common repository. These aren't being tracked on your fork.

stanbasickes commented 4 years ago

Phil: I apologize for the posting. The equations being used in the app.py are wrong. I was trying to bring that to the attention of the modelers. However, I cannot figure out how to properly use github to do that.

Stan Basickes

On Thu, Apr 2, 2020 at 6:41 PM Phil Miller notifications@github.com wrote:

Closed #455 https://github.com/CodeForPhilly/chime/issues/455.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/CodeForPhilly/chime/issues/455#event-3194175285, or unsubscribe https://github.com/notifications/unsubscribe-auth/AO4T4NXFYOWG2SGXX3YD353RKUIC5ANCNFSM4L24DBBQ .

PhilMiller commented 4 years ago

The equations used in the model do not follow the mechanics one would naively expect, but they are long-established practice in epidemiology. More importantly, the model as it exists is validated as actually having its forecasts matching observations at several hospital systems.

For the still-open issues where you actually posted your code, analysis, and results, I expect one of the professionals at Penn will have a look.

stanbasickes commented 4 years ago

I hope they do !!

Stan Basickes, PhD Statistics U of P 1972

On Fri, Apr 3, 2020 at 11:38 AM Phil Miller notifications@github.com wrote:

The equations used in the model do not follow the mechanics one would naively expect, but they are long-established practice in epidemiology. More importantly, the model as it exists is validated as actually having its forecasts matching observations at several hospital systems.

For the still-open issues where you actually posted your code, analysis, and results, I expect one of the professionals at Penn will have a look.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/CodeForPhilly/chime/issues/455#issuecomment-608511356, or unsubscribe https://github.com/notifications/unsubscribe-auth/AO4T4NUZISDJILKBFOMXEU3RKX7G3ANCNFSM4L24DBBQ .