Closed dmitra79 closed 4 years ago
Ah, there was a divide by zero bug in the inference code. This should be fixed now!
I checked out the latest version. The problem is resolved for 3 weak labels, but not for 2, ie. the following still produces nans:
n=100
a=np.ones(n)
b=-1*np.ones(2*n)
z=np.concatenate([a, b, b, a]).reshape((2,3*n)).transpose()
print(z.shape)
squid_run(z)
Is the method completely not applicable when there are only 2 weak labels?
Yes, we need at least three (conditionally-independent) labeling functions to run.
There's now a check for this on label model creation!
Great - thank you!
Hello,
I've found that in a number of situations
predict_proba_marginalized
returns 'nan'. I didn't see this behavior in tutorials or documented, and wasn't sure how to interpret it. Here is one example:The first of the runs returns 'nan' for all instance probabilities (and 1s for estimated probabilities). The second run returns:
The same runs with only two weak labels, ie: ''' z=np.concatenate([a, b, b, a]).reshape((2,2*n)).transpose() ''' result in 'nan' in both cases.