Using scan to loop across differenet length of vectors

Questions
,
7

Thanks for the help.
Lets assume this is the original model:

with pm.Model() as m:

alpha = pm.Beta('alpha', 1,1, shape=n_subj)
beta = pm.Normal('beta',0, 1, shape=n_subj)
eps = pm.HalfNormal('eps', 5)

Qs = 0.5 * tt.ones((n_subj,2), dtype='float64') # set values for boths stimuli (CS+, CS-)
vec = 0.5 * tt.ones((n_subj,1), dtype='float64') # vector to save the relevant stimulus's expactation

[Qs,vec, pe], updates = theano.scan(
    fn=update_Q,
    sequences=[stim, shock],
    outputs_info=[Qs, vec, None],
    non_sequences=[alpha, n_subj])


vec_ = vec[trials, subj,0] * beta[subj]

scrs = pm.Normal('scrs', mu = vec_, sd = eps, observed=scrMat) 

# add matrix of expected values (trials X subjects)
ev = pm.Deterministic('expected_value', vec_)
# add PE
pe = pm.Deterministic('pe', pe)

tr = pm.sample(target_accept=.9, chains=4, cores=10, return_inferencedata=True)

the function receives matrices of stim and shock (ntrials X subjects).
Later on, I use a similar structure for the scrMat (observed variable to fit, SCR with ntrials X nSubs as well).

So, as long as the matrices have the same size, I can send each of the three options of matrices to the same scan, just at different times. Next, I need to fit everything with the likelihood function, I can also split it into three? wouldn’t it be as if a just run each set of subjects separately?