How to use arviz - 10 common examples

ax[j, i].tick_params(labelsize=xt_labelsize)
                ax[j, i].set_title(
                    "{} - {}".format(models[i], models[j + 1]), fontsize=titlesize, wrap=True
                )
        if xlabels:
            set_xticklabels(ax[-1, -1], coord_labels)
            fig.autofmt_xdate()
            fig.tight_layout()
        if legend:
            ncols = len(handles) // 6 + 1
            ax[0, 1].legend(
                handles=handles, ncol=ncols, title=color, bbox_to_anchor=(0, 1), loc="upper left"
            )

    if backend_show(show):
        plt.show()

    return ax

"""
LOO-PIT Overlay Plot
====================

_thumb: .5, .7
"""
import arviz as az

az.style.use("arviz-darkgrid")

idata = az.load_arviz_data("non_centered_eight")

az.plot_loo_pit(idata=idata, y="obs", color="indigo")

"""
LOO-PIT ECDF Plot
=================

_thumb: .5, .7
"""
import arviz as az

az.style.use("arviz-darkgrid")

idata = az.load_arviz_data("radon")
log_like = idata.sample_stats.log_likelihood.sel(chain=0).values.T
log_weights = az.psislw(-log_like)[0]

az.plot_loo_pit(idata, y="y_like", log_weights=log_weights, ecdf=True, color="maroon")

chain_data = []
            for path in posterior_predictive:
                parsed_output = _read_output(path)
                for sample, *_ in parsed_output:
                    chain_data.append(sample)
            data = _unpack_dataframes(chain_data)
        else:
            if isinstance(posterior_predictive, str):
                posterior_predictive = [posterior_predictive]
            posterior_predictive_cols = [
                col
                for col in columns
                if any(item == col.split(".")[0] for item in posterior_predictive)
            ]
            data = _unpack_dataframes([item[posterior_predictive_cols] for item in self.posterior])
        return dict_to_dataset(data, coords=self.coords, dims=self.dims)

def prior_predictive_to_xarray(self):
        """Convert prior_predictive samples to xarray."""
        prior_predictive = self.prior_predictive
        columns = self.prior.column_names

        if isinstance(prior_predictive, str):
            prior_predictive = [prior_predictive]
        valid_cols = [col for col in columns if col.split(".")[0] in set(prior_predictive)]
        data = _unpack_frame(self.prior.sample, columns, valid_cols)
        return dict_to_dataset(data, library=self.cmdstanpy, coords=self.coords, dims=self.dims)

def posterior_predictive_to_xarray(self):
        """Convert posterior_predictive samples to xarray."""
        posterior_predictive = self.posterior_predictive
        columns = self.posterior.column_names

        if isinstance(posterior_predictive, str):
            posterior_predictive = [posterior_predictive]
        valid_cols = [col for col in columns if col.split(".")[0] in set(posterior_predictive)]
        data = _unpack_frame(self.posterior.sample, columns, valid_cols)
        return dict_to_dataset(data, library=self.cmdstanpy, coords=self.coords, dims=self.dims)

"""
LOO-PIT Overlay Plot
====================

_thumb: .5, .7
"""
import arviz as az

az.style.use("arviz-darkgrid")

idata = az.load_arviz_data("non_centered_eight")

az.plot_loo_pit(idata=idata, y="obs", color="indigo")

    @requires("prior")
    def prior_to_xarray(self):
        """Convert prior samples to xarray."""
        # filter prior_predictive
        prior_predictive = self.prior_predictive
        columns = self.prior[0].columns
        if prior_predictive is None or (
            isinstance(prior_predictive, str) and prior_predictive.lower().endswith(".csv")
        ):
            prior_predictive = []
        elif isinstance(prior_predictive, str):
            prior_predictive = [col for col in columns if prior_predictive == col.split(".")[0]]
        else:
            prior_predictive = [
                col
                for col in columns
                if any(item == col.split(".")[0] for item in prior_predictive)

    @requires("posterior_predictive")
    def posterior_predictive_to_xarray(self):
        """Convert posterior_predictive samples to xarray."""
        posterior_predictive = self.posterior_predictive
        columns = self.posterior.column_names

        if isinstance(posterior_predictive, str):
            posterior_predictive = [posterior_predictive]
        valid_cols = [col for col in columns if col.split(".")[0] in set(posterior_predictive)]
        data = _unpack_frame(self.posterior.sample, columns, valid_cols)
        return dict_to_dataset(data, library=self.cmdstanpy, coords=self.coords, dims=self.dims)

    @requires("sample_stats")
    def sample_stats_to_xarray(self):
        """Extract sample_stats from fit."""
        dtypes = {"divergent__": bool, "n_leapfrog__": np.int64, "treedepth__": np.int64}

        # copy dims and coords
        dims = deepcopy(self.dims) if self.dims is not None else {}
        coords = deepcopy(self.coords) if self.coords is not None else {}

        sampler_params = self.sample_stats
        log_likelihood = self.log_likelihood
        if isinstance(log_likelihood, str):
            log_likelihood_cols = [
                col for col in self.posterior[0].columns if log_likelihood == col.split(".")[0]
            ]
            log_likelihood_vals = [item[log_likelihood_cols] for item in self.posterior]