I’m trying to use a pymc model to impute missing categorical values in my dataset. When I set the observed data in a Categorical distribution to a masked array: data = np.ma.masked_equal([1, 1, 0, 0, 2, -1, -1], -1) with pm.Model(): idx = pm.Categorical(f"{var}_idx", p=[0.1, 0.2, 0.7], observe…

Trying to impute missing categorical data

ricardoV94 January 2, 2023, 5:49pm 4

Looks like a bug. As a short solution you can cast the result variable

idx = idx.astype("float64")

Do you mind opening an issue in the GitHub repo?

Topic		Replies	Views
Trouble specificying X \| a, b, c, d ~ Categorical( . ) Questions	5	508	March 2, 2019
pm.Categorical behaves differently in a model versus as pm.Categorical.dist Questions	2	929	August 8, 2018
Categorical variable with p equal to deterministic that depends on observed categorical Questions	2	782	February 5, 2018
Strange error with Categorical distribution Questions	3	488	August 15, 2018
Marginalizing over missing categories Questions	1	726	June 17, 2020