I am using a MatrixNormal distribution in my model:
df contains incomplete data, as a lot of its entries are
None. As suggested by https://nbviewer.jupyter.org/github/fonnesbeck/scipy2014_tutorial/blob/master/3_Introduction-to-PyMC.ipynb, the model ignores the missing obseravtions and performs inference. However, when I use
pm.traceplot(trace), I get the following error:
log_likelihood_vals_point log_like_val = log_like_val[~var.observations.mask] IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed
This happens even if I define the
var_names of the traceplot and exclude
Any ideas on how to fix this and plot the trace?