"""High level conversion functions."""
import numpy as np
import xarray as xr
try:
from tree import is_nested
except ImportError:
is_nested = lambda obj: False
from .base import dict_to_dataset
from .inference_data import InferenceData
from .io_beanmachine import from_beanmachine
from .io_cmdstan import from_cmdstan
from .io_cmdstanpy import from_cmdstanpy
from .io_emcee import from_emcee
from .io_numpyro import from_numpyro
from .io_pyro import from_pyro
from .io_pystan import from_pystan
# pylint: disable=too-many-return-statements
[docs]
def convert_to_inference_data(obj, *, group="posterior", coords=None, dims=None, **kwargs):
r"""Convert a supported object to an InferenceData object.
This function sends `obj` to the right conversion function. It is idempotent,
in that it will return arviz.InferenceData objects unchanged.
Parameters
----------
obj : dict, str, np.ndarray, xr.Dataset, pystan fit
A supported object to convert to InferenceData:
| InferenceData: returns unchanged
| str: Attempts to load the cmdstan csv or netcdf dataset from disk
| pystan fit: Automatically extracts data
| cmdstanpy fit: Automatically extracts data
| cmdstan csv-list: Automatically extracts data
| emcee sampler: Automatically extracts data
| pyro MCMC: Automatically extracts data
| beanmachine MonteCarloSamples: Automatically extracts data
| xarray.Dataset: adds to InferenceData as only group
| xarray.DataArray: creates an xarray dataset as the only group, gives the
array an arbitrary name, if name not set
| dict: creates an xarray dataset as the only group
| numpy array: creates an xarray dataset as the only group, gives the
array an arbitrary name
group : str
If `obj` is a dict or numpy array, assigns the resulting xarray
dataset to this group. Default: "posterior".
coords : dict[str, iterable]
A dictionary containing the values that are used as index. The key
is the name of the dimension, the values are the index values.
dims : dict[str, List(str)]
A mapping from variables to a list of coordinate names for the variable
kwargs
Rest of the supported keyword arguments transferred to conversion function.
Returns
-------
InferenceData
"""
kwargs[group] = obj
kwargs["coords"] = coords
kwargs["dims"] = dims
# Cases that convert to InferenceData
if isinstance(obj, InferenceData):
if coords is not None or dims is not None:
raise TypeError("Cannot use coords or dims arguments with InferenceData value.")
return obj
elif isinstance(obj, str):
if obj.endswith(".csv"):
if group == "sample_stats":
kwargs["posterior"] = kwargs.pop(group)
elif group == "sample_stats_prior":
kwargs["prior"] = kwargs.pop(group)
return from_cmdstan(**kwargs)
else:
if coords is not None or dims is not None:
raise TypeError(
"Cannot use coords or dims arguments reading InferenceData from netcdf."
)
return InferenceData.from_netcdf(obj)
elif (
obj.__class__.__name__ in {"StanFit4Model", "CmdStanMCMC"}
or obj.__class__.__module__ == "stan.fit"
):
if group == "sample_stats":
kwargs["posterior"] = kwargs.pop(group)
elif group == "sample_stats_prior":
kwargs["prior"] = kwargs.pop(group)
if obj.__class__.__name__ == "CmdStanMCMC":
return from_cmdstanpy(**kwargs)
else: # pystan or pystan3
return from_pystan(**kwargs)
elif obj.__class__.__name__ == "EnsembleSampler": # ugly, but doesn't make emcee a requirement
return from_emcee(sampler=kwargs.pop(group), **kwargs)
elif obj.__class__.__name__ == "MonteCarloSamples":
return from_beanmachine(sampler=kwargs.pop(group), **kwargs)
elif obj.__class__.__name__ == "MCMC" and obj.__class__.__module__.startswith("pyro"):
return from_pyro(posterior=kwargs.pop(group), **kwargs)
elif obj.__class__.__name__ == "MCMC" and obj.__class__.__module__.startswith("numpyro"):
return from_numpyro(posterior=kwargs.pop(group), **kwargs)
# Cases that convert to xarray
if isinstance(obj, xr.Dataset):
dataset = obj
elif isinstance(obj, xr.DataArray):
if obj.name is None:
obj.name = "x"
dataset = obj.to_dataset()
elif isinstance(obj, dict):
dataset = dict_to_dataset(obj, coords=coords, dims=dims)
elif is_nested(obj) and not isinstance(obj, (list, tuple)):
dataset = dict_to_dataset(obj, coords=coords, dims=dims)
elif isinstance(obj, np.ndarray):
dataset = dict_to_dataset({"x": obj}, coords=coords, dims=dims)
elif isinstance(obj, (list, tuple)) and isinstance(obj[0], str) and obj[0].endswith(".csv"):
if group == "sample_stats":
kwargs["posterior"] = kwargs.pop(group)
elif group == "sample_stats_prior":
kwargs["prior"] = kwargs.pop(group)
return from_cmdstan(**kwargs)
else:
allowable_types = (
"xarray dataarray",
"xarray dataset",
"dict",
"pytree (if 'dm-tree' is installed)",
"netcdf filename",
"numpy array",
"pystan fit",
"emcee fit",
"pyro mcmc fit",
"numpyro mcmc fit",
"cmdstan fit csv filename",
"cmdstanpy fit",
)
raise ValueError(
f'Can only convert {", ".join(allowable_types)} to InferenceData, '
f"not {obj.__class__.__name__}"
)
return InferenceData(**{group: dataset})
[docs]
def convert_to_dataset(obj, *, group="posterior", coords=None, dims=None):
"""Convert a supported object to an xarray dataset.
This function is idempotent, in that it will return xarray.Dataset functions
unchanged. Raises `ValueError` if the desired group can not be extracted.
Note this goes through a DataInference object. See `convert_to_inference_data`
for more details. Raises ValueError if it can not work out the desired
conversion.
Parameters
----------
obj : dict, str, np.ndarray, xr.Dataset, pystan fit
A supported object to convert to InferenceData:
- InferenceData: returns unchanged
- str: Attempts to load the netcdf dataset from disk
- pystan fit: Automatically extracts data
- xarray.Dataset: adds to InferenceData as only group
- xarray.DataArray: creates an xarray dataset as the only group, gives the
array an arbitrary name, if name not set
- dict: creates an xarray dataset as the only group
- numpy array: creates an xarray dataset as the only group, gives the
array an arbitrary name
group : str
If `obj` is a dict or numpy array, assigns the resulting xarray
dataset to this group.
coords : dict[str, iterable]
A dictionary containing the values that are used as index. The key
is the name of the dimension, the values are the index values.
dims : dict[str, List(str)]
A mapping from variables to a list of coordinate names for the variable
Returns
-------
xarray.Dataset
"""
inference_data = convert_to_inference_data(obj, group=group, coords=coords, dims=dims)
dataset = getattr(inference_data, group, None)
if dataset is None:
raise ValueError(
"Can not extract {group} from {obj}! See {filename} for other "
"conversion utilities.".format(group=group, obj=obj, filename=__file__)
)
return dataset
