I have been stuck in a Problem, for more than 2 weeks now.

I need to generate data set, that have actual Probability.

I am doing an experiment, my ML framework gives me probability intervals for the class labels (each sample). I want to check how many times the "actual Probability" is included in the Intervall.

I have tried Uniform distribution, and I have generated in the range [0,1) 4-dimensional data. I have summed it together and divided it by the number of the features. That's my actual probability. The actual probability is given to the Bernoulli distribution to get a class labels. But The actual probabilities are also uniform distribution, and it is supposed to. But Framework does not learn, almost all prediction values are 1. Example: I have created dataset using uniform distribution. 4 features + target

0.5507979, 0.93330316, 0.9868236, 0.533589, 1
0.2909047, 0.63284733, 0.6656879, 0.174750, 0

actual probability: (0.5507979+0.93330316+0.9868236+0.533589)/4 = 0.7511284

actual probability (0.7511284) given to Bernoulli distribution, gives me class label 1.

I have an idea that if I generate first the True Probability and according to that create feature values but I do not know how to do it. Example: let's say True probability (p) = 0.80, and most probably Bernoulli gives me a class label 1. Now how to generate some features so I can give them to ML Framework and compute the Intervall.

I am using Python.

your support will be highly appreciated.

Thank you very much in Advance.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.