How to create Bayesian data fusion in python with pymc3?

Questions v3
2

Other possibilities of having different results in the bayesian fusion:

  • Could it be because I am using weakly priors in the pymc3 models where the weakly priors would cause the bayesian fusion to have different results, everytime I re-run the code?
  • Could it be because the assigned draws and tune parameters in the trace code? as I assign the trace code to,
    trace = pm.sample(draws=1000, tune=500,......)
    Which would possible lead to different prior values?