# Transitioning from pystan 2.x to 3.x

**URL:** <https://discourse.mc-stan.org/t/transitioning-from-pystan-2-x-to-3-x/33468>\
**Category:** PyStan\
**Created:** [November 29, 2023, 12:09pm UTC](https://discourse.mc-stan.org/t/transitioning-from-pystan-2-x-to-3-x/33468 "2023-11-29T12:09:13Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Garren\_Hermanus](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/garren_hermanus/32/17356_2.png) [@Garren\_Hermanus](https://discourse.mc-stan.org/u/Garren_Hermanus)\
**Post date:** [November 29, 2023, 12:09pm UTC](https://discourse.mc-stan.org/t/transitioning-from-pystan-2-x-to-3-x/33468/1 "2023-11-29T12:09:13Z")

</div>

Hey all, I am fairly new to stan (started using it this year only). I have for the majority of the year used pystan 2.19, but have transitioned to pystan 3.x. I just need some clarity on the differences in function calls, and specifying additional parameters such as adapt\_delta and max\_treedepth. I typically got false divergences for my models, so I always made my adapt\_delta = 0.99. I also saved information like my stepsize and inv\_metric to do additional runs if required.

I have familiarized myself with the basics of 3.x, like instead of pystan.StanModel() we use stan.build() and model.sampling() we use model.sample()

In pystan 2.19 adapat\_delta and max\_treedepth is accessed through the control dict variable. I just want to know:

1. How I can change these variables in 3.x?
2. How do I access the inv\_metric and stepsize after adaption warmup finishes and specify these for new additional runs?
3. How do I access the log posterior ‘lp\_\_’ for each sample?
4. How do I get convergence metrics, rhat and neff?

I’ve read the documentation for pystan3.7 but could not find any information regarding these

---

<div class="post-metadata">

**Author:** ![ahartikainen](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/ahartikainen/32/321_2.png) [@ahartikainen](https://discourse.mc-stan.org/u/ahartikainen)\
**Post date:** [November 29, 2023, 7:29pm UTC](https://discourse.mc-stan.org/t/transitioning-from-pystan-2-x-to-3-x/33468/2 "2023-11-29T19:29:41Z")

</div>

Hi, yes there are couple of different changes in PyStan 3 vs PyStan 2. Mainly that PyStan 3 has support for main functionality and it has dropped support for all the experimental stuff.

If you need to use these properties I recommend using CmdStanPy which has larger support for experimental properties.

> [@Garren\_Hermanus](#):
>
> - How I can change these variables in 3.x?
> - How do I access the inv\_metric and stepsize after adaption warmup finishes and specify these for new additional runs?
> - How do I access the log posterior ‘lp\_\_’ for each sample?
> - How do I get convergence metrics, rhat and neff?

1. Pystan uses similar keywords as does CmdStan, see [https://github.com/stan-dev/httpstan/blob/main/httpstan/services/cmdstan-help-all.json](https://github.com/stan-dev/httpstan/blob/main/httpstan/services/cmdstan-help-all.json)

2. I recommend CmdStanPy for inverse metric. `fit["stepsize__"]`. Also arviz inferencedata has support for these.

3. `fit["lp__"]`

4. I recommend using ArviZ [arviz.from\_pystan — ArviZ 0.16.1 documentation](https://python.arviz.org/en/stable/api/generated/arviz.from_pystan.html)

I recommend also adding named dimensions and coordinates where suitable and then saving the results as netcdf4.

```python
import arviz as az
inferencedata = az.from_pystan(posterior=fit, posterior_model=model)
summary = az.summary(inferencedata)
print(summary)

```

---

<div class="post-metadata">

**Author:** ![Garren\_Hermanus](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/garren_hermanus/32/17356_2.png) [@Garren\_Hermanus](https://discourse.mc-stan.org/u/Garren_Hermanus)\
**Post date:** [November 29, 2023, 11:12pm UTC](https://discourse.mc-stan.org/t/transitioning-from-pystan-2-x-to-3-x/33468/3 "2023-11-29T23:12:12Z")

</div>

Thanks @ahartikainen. Looking at the cmdstan-help-all.json file cleared up a lot of confusion I had. Just to clarify a few aspects.

1. I plotted the stepsize and from this I concluded that the stepsizes stores `num_chains*num_samples` elements but only the first `num_chains` is the stepsizes, the rest are repeats? And also by this logic this means that the posterior samples in the fit dict are from chains: 1, 2, 3, 4, 1, 2, 3, 4, … and not from chains: 1, 1, …, 2, 2, …, 3, 3, …, 4, 4, … as with pystan 2.19 fit.extract() method?

2. I see that the cmdstan-help-all.json has an `engaged` parameters. If I specify `num_warmup=0` do I still need to specify `engaged=False` (for additional runs without adaption)? When doing this I get an error stating “ValueError: {‘json’: {‘engaged’: [‘Unknown field.’]}}”

3. Trying to set `metric=np.array([1, 1, 1])` (for a simple linear regression problem with slope intercept and variance as model parameters, just to check if I understand the interface correctly), I again get a value “ValueError: {‘json’: {‘metric’: [‘Unknown field.’]}}”

4. Q2-3 leads into my final question. When you say you recommend using cmdstanpy is this only for extracting the metric or is it for drawing additional samples from an already tuned fit? I.e. using the metric and stepsize, with some init (probably sample means) from the fit as fixed parameters?

---

<div class="post-metadata">

**Author:** ![mitzimorris](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/mitzimorris/32/29_2.png) [@mitzimorris](https://discourse.mc-stan.org/u/mitzimorris)\
**Post date:** [November 30, 2023, 1:09pm UTC](https://discourse.mc-stan.org/t/transitioning-from-pystan-2-x-to-3-x/33468/4 "2023-11-30T13:09:41Z")

</div>

> [@Garren\_Hermanus](#):
>
> When you say you recommend using cmdstanpy is this only for extracting the metric or is it for drawing additional samples from an already tuned fit? I.e. using the metric and stepsize, with some init (probably sample means) from the fit as fixed parameters?

there is a nice summary of the differences here: [CmdStanPy 1.0 - #3 by WardBrian](https://discourse.mc-stan.org/t/cmdstanpy-1-0/25552/3)