Will have to leave this to others, since I have both insufficient machine learning expertise to comment critically, as well as having a potential conflict of interest related to affiliations.
-
-
-
Fair enough!
End of conversation
New conversation -
-
-
Wouldn’t call it ML - it’s more of a sample-simulate-adjust parameters loop with some parameters being fixed by authors (eg durations of infective periods) and others assumed to follow specific distributions (that authors do not justif, safe for Student-t trick)
-
Since it’s pretty much garbage in-garbage out, focusing on the data they use (eg equating symptomatic cases to diagnosed confirmed ones) gives you an upper bound on model performance. Specifics would need source code examination, which should be doable given they seem to use .py
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.