Question on Conditional Distributions in Tensorflow_Probability.

[Gaurav Joshi]

I wanted to know about the features available in tensorflow_probability for making conditional Distributions. More specifically how can I make a distribution which is conditioned on another random variable.

My use case is: I have to pick to times (t_1 and t_2) from t = 0 to 10s, but they have to be picked in a beta distribution, ie t_1 is a beta distribution from 0 to 10s, and t_2 is a beta distribution from t_1 to 10s. For this, I have to make a t_2 as a conditional beta distribution conditioned on t_1.

 If anyone can point me out to some docs or show me how the conditional beta can be made, I will be very grateful.

[rif]

The general term for this in TensorFlow Probability is "JointDistribution". A couple of relevant tutorials:

https://github.com/tensorflow/probability/blob/main/tensorflow_probability/examples/jupyter_notebooks/Modeling_with_JointDistribution.ipynb

https://github.com/tensorflow/probability/blob/main/tensorflow_probability/examples/jupyter_notebooks/JointDistributionAutoBatched_A_Gentle_Tutorial.ipynb

Note that JointDistribution comes in a few different "flavors" depending on your syntactic preferences.

Hope this helps.

Best,

rif

--
You received this message because you are subscribed to the Google Groups "TensorFlow Probability" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tfprobabilit...@tensorflow.org.
To view this discussion on the web visit https://groups.google.com/a/tensorflow.org/d/msgid/tfprobability/17f6b6f5-4ed2-4933-8be7-d0d1685fda21n%40tensorflow.org.