I’m following this tutorial on accelerated failure time (AFT) modeling and I’m not sure how to extract shape and scale parameters for the Weibull AFT model. To simplify, suppose there are not covariates. If I were to use lifelines’ WeibullAFTFitter or survival's survreg, I’d estimate two parameters: shape and scale. In the pymc3 tutorial, is the shape related to s, while the scale is related to beta? See these notes on survreg's parameterization.