search

tapios / risk-networks

Code for risk networks: a blend of compartmental models, graphs, data assimilation and semi-supervised learning
Other
2 stars 2 forks source link

observe takes noisy_measurement as an argument but never uses it #114

Closed dburov190 closed 4 years ago

dburov190 commented 4 years ago

In two places, here: https://github.com/dburov190/risk-networks/blob/8d8bbfc0b61621d9ef48df22b33251dd8f386b64/epiforecast/measurements.py#L325-L330 and here: https://github.com/dburov190/risk-networks/blob/8d8bbfc0b61621d9ef48df22b33251dd8f386b64/epiforecast/measurements.py#L424-L429

noisy_measurement is being passed into two observe methods as an argument; again, is this okay to delete or something actually should take place?

odunbar commented 4 years ago

We will use it, just not during our current testing. It performs a random flip of a test result, in accordance with the amount of noise we have on the measurement

dburov190 commented 4 years ago

alright, fair enough, but let's keep this issue open until it's resolved

odunbar commented 4 years ago

Ok i see why this is an issue, what you really want is the variable to be in the constructor.