Overflow problem

Questions
1

Hi,everyone，my Bayesian model encountered an overflow problem,

I don’t know how to solve it. any help will be very grateful,my code is as follows:

with pm.Model() as model:
    BoundNormal = pm.Bound(pm.Normal, lower=0., upper=1.)
    sigma_w = pm.InverseGamma('sigma_w',1.0,0.1)
    rho = BoundNormal('rho', mu=0.4, sd=0.01)
    cov0=sigma_w**2*tt.pow(rho,H0)

    omega=pm.MvNormal('omega',mu=mu0,cov=cov0,shape=(1,n))    

    mu1=pm.Normal('mu1',mu=0.0,sd=0.01)
    xi=pm.InverseGamma('xi',11.0,0.1)
    b=pm.Normal('b', mu=0.0, sd=tt.sqrt(xi),shape=(N,1))

    mu2=(mu1+b)*e+omega
    sigma=pm.HalfNormal('sigma',0.01)

    y=pm.Normal('y',mu=mu2,sd=sigma,observed=dataSet2)  

with model:
    trace=pm.sample(n=2000,njobs=1)

error occurred when sampling was made up to 40%

raise ValueError('Mass matrix contains zeros on the diagonal. '

ValueError: Mass matrix contains zeros on the diagonal. Some derivatives might always be zero.
2

If you do a search on the discourse, there are quite a few idea of how to debug it. In general, it is cause by improper prior.

3

Thank you for your guidance. I will search related questions to try to improve my own priori.