Source code for pymc.model_graph

#   Copyright 2024 The PyMC Developers
#
#   Licensed under the Apache License, Version 2.0 (the "License");
#   you may not use this file except in compliance with the License.
#   You may obtain a copy of the License at
#
#       http://www.apache.org/licenses/LICENSE-2.0
#
#   Unless required by applicable law or agreed to in writing, software
#   distributed under the License is distributed on an "AS IS" BASIS,
#   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#   See the License for the specific language governing permissions and
#   limitations under the License.
import warnings

from collections import defaultdict
from collections.abc import Callable, Iterable
from dataclasses import dataclass
from enum import Enum
from os import path
from typing import Any, cast

from pytensor import function
from pytensor.graph import Apply
from pytensor.graph.basic import ancestors, walk
from pytensor.scalar.basic import Cast
from pytensor.tensor.elemwise import Elemwise
from pytensor.tensor.random.op import RandomVariable
from pytensor.tensor.shape import Shape
from pytensor.tensor.variable import TensorVariable

import pymc as pm

from pymc.util import VarName, get_default_varnames, get_var_name

__all__ = (
    "ModelGraph",
    "model_to_graphviz",
    "model_to_networkx",
)


@dataclass
class DimInfo:
    names: tuple[str | None, ...]
    lengths: tuple[int, ...]

    def __post_init__(self) -> None:
        if len(self.names) != len(self.lengths):
            raise ValueError("The number of names and lengths must be equal.")

    def __hash__(self):
        return hash((self.names, self.lengths))

    def __bool__(self) -> bool:
        return len(self.lengths) > 0 or len(self.names) > 0


PlateLabelFunc = Callable[[DimInfo], str]


def create_plate_label_without_dim_length(
    dim_info: DimInfo,
) -> str:
    return " x ".join(
        f"{dname}" if dname else f"{dlen}"
        for (dname, dlen) in zip(dim_info.names, dim_info.lengths)
    )


def create_plate_label_with_dim_length(
    dim_info: DimInfo,
) -> str:
    return " x ".join(
        f"{dname} ({dlen})" if dname else f"{dlen}"
        for (dname, dlen) in zip(dim_info.names, dim_info.lengths)
    )


def fast_eval(var):
    return function([], var, mode="FAST_COMPILE")()


class NodeType(str, Enum):
    """Enum for the types of nodes in the graph."""

    POTENTIAL = "Potential"
    FREE_RV = "Free Random Variable"
    OBSERVED_RV = "Observed Random Variable"
    DETERMINISTIC = "Deterministic"
    DATA = "Data"


@dataclass
class NodeInfo:
    var: TensorVariable
    node_type: NodeType

    def __hash__(self):
        return hash(self.var.name)


@dataclass
class Plate:
    dim_info: DimInfo
    variables: list[NodeInfo]

    def __eq__(self, other) -> bool:
        if not isinstance(other, Plate):
            return False

        return self.dim_info == other.dim_info and set(self.variables) == set(other.variables)


GraphvizNodeKwargs = dict[str, Any]
NodeFormatter = Callable[[TensorVariable], GraphvizNodeKwargs]


def default_potential(var: TensorVariable) -> GraphvizNodeKwargs:
    """Default data for potential in the graph."""
    return {
        "shape": "octagon",
        "style": "filled",
        "label": f"{var.name}\n~\nPotential",
    }


def random_variable_symbol(var: TensorVariable) -> str:
    """Get the symbol of the random variable."""
    symbol = var.owner.op.__class__.__name__

    if symbol.endswith("RV"):
        symbol = symbol[:-2]

    return symbol


def default_free_rv(var: TensorVariable) -> GraphvizNodeKwargs:
    """Default data for free RV in the graph."""
    symbol = random_variable_symbol(var)

    return {
        "shape": "ellipse",
        "style": None,
        "label": f"{var.name}\n~\n{symbol}",
    }


def default_observed_rv(var: TensorVariable) -> GraphvizNodeKwargs:
    """Default data for observed RV in the graph."""
    symbol = random_variable_symbol(var)

    return {
        "shape": "ellipse",
        "style": "filled",
        "label": f"{var.name}\n~\n{symbol}",
    }


def default_deterministic(var: TensorVariable) -> GraphvizNodeKwargs:
    """Default data for the deterministic in the graph."""
    return {
        "shape": "box",
        "style": None,
        "label": f"{var.name}\n~\nDeterministic",
    }


def default_data(var: TensorVariable) -> GraphvizNodeKwargs:
    """Default data for the data in the graph."""
    return {
        "shape": "box",
        "style": "rounded, filled",
        "label": f"{var.name}\n~\nData",
    }


def get_node_type(var_name: VarName, model) -> NodeType:
    """Return the node type of the variable in the model."""
    v = model[var_name]

    if v in model.deterministics:
        return NodeType.DETERMINISTIC
    elif v in model.free_RVs:
        return NodeType.FREE_RV
    elif v in model.observed_RVs:
        return NodeType.OBSERVED_RV
    elif v in model.data_vars:
        return NodeType.DATA
    else:
        return NodeType.POTENTIAL


NodeTypeFormatterMapping = dict[NodeType, NodeFormatter]

DEFAULT_NODE_FORMATTERS: NodeTypeFormatterMapping = {
    NodeType.POTENTIAL: default_potential,
    NodeType.FREE_RV: default_free_rv,
    NodeType.OBSERVED_RV: default_observed_rv,
    NodeType.DETERMINISTIC: default_deterministic,
    NodeType.DATA: default_data,
}


def update_node_formatters(node_formatters: NodeTypeFormatterMapping) -> NodeTypeFormatterMapping:
    node_formatters = {**DEFAULT_NODE_FORMATTERS, **node_formatters}

    unknown_keys = set(node_formatters.keys()) - set(NodeType)
    if unknown_keys:
        raise ValueError(
            f"Node formatters must be of type NodeType. Found: {list(unknown_keys)}."
            f" Please use one of {[node_type.value for node_type in NodeType]}."
        )

    return node_formatters


AddNode = Callable[[str, GraphvizNodeKwargs], None]


def _make_node(
    node: NodeInfo,
    *,
    node_formatters: NodeTypeFormatterMapping,
    add_node: AddNode,
    cluster: str | None = None,
    formatting: str = "plain",
):
    """Attaches the given variable to a graphviz or networkx Digraph"""
    node_formatter = node_formatters[node.node_type]
    kwargs = node_formatter(node.var)

    if cluster is not None:
        kwargs["cluster"] = cluster

    var_name: str = cast(str, node.var.name)
    add_node(var_name.replace(":", "&"), **kwargs)  # type: ignore


class ModelGraph:
    def __init__(self, model):
        self.model = model
        self._all_var_names = get_default_varnames(self.model.named_vars, include_transformed=False)
        self.var_list = self.model.named_vars.values()

    def get_parent_names(self, var: TensorVariable) -> set[VarName]:
        if var.owner is None or var.owner.inputs is None:
            return set()

        def _filter_non_parameter_inputs(var):
            node = var.owner
            if isinstance(node.op, Shape):
                # Don't show shape-related dependencies
                return []
            if isinstance(node.op, RandomVariable):
                # Filter out rng and size parameters or RandomVariable nodes
                return node.op.dist_params(node)
            else:
                # Otherwise return all inputs
                return node.inputs

        blockers = set(self.model.named_vars)

        def _expand(x):
            nonlocal blockers
            if x.name in blockers:
                return [x]
            if isinstance(x.owner, Apply):
                return reversed(_filter_non_parameter_inputs(x))
            return []

        parents = set()
        for x in walk(nodes=_filter_non_parameter_inputs(var), expand=_expand):
            # Only consider nodes that are in the named model variables.
            vname = getattr(x, "name", None)
            if isinstance(vname, str) and vname in self._all_var_names:
                parents.add(VarName(vname))

        return parents

    def vars_to_plot(self, var_names: Iterable[VarName] | None = None) -> list[VarName]:
        if var_names is None:
            return self._all_var_names

        selected_names = set(var_names)

        # .copy() because sets cannot change in size during iteration
        for var_name in selected_names.copy():
            if var_name not in self._all_var_names:
                raise ValueError(f"{var_name} is not in this model.")

            for model_var in self.var_list:
                if model_var in self.model.observed_RVs:
                    if self.model.rvs_to_values[model_var] == self.model[var_name]:
                        selected_names.add(model_var.name)

        selected_ancestors = set(
            filter(
                lambda rv: rv.name in self._all_var_names,
                list(ancestors([self.model[var_name] for var_name in selected_names])),
            )
        )

        for var in selected_ancestors.copy():
            if var in self.model.observed_RVs:
                selected_ancestors.add(self.model.rvs_to_values[var])

        # ordering of self._all_var_names is important
        return [get_var_name(var) for var in selected_ancestors]

    def make_compute_graph(
        self, var_names: Iterable[VarName] | None = None
    ) -> dict[VarName, set[VarName]]:
        """Get map of var_name -> set(input var names) for the model"""
        input_map: dict[VarName, set[VarName]] = defaultdict(set)

        for var_name in self.vars_to_plot(var_names):
            var = self.model[var_name]
            parent_name = self.get_parent_names(var)
            input_map[var_name] = input_map[var_name].union(parent_name)

            if var in self.model.observed_RVs:
                obs_node = self.model.rvs_to_values[var]

                # loop created so that the elif block can go through this again
                # and remove any intermediate ops, notably dtype casting, to observations
                while True:
                    obs_name = obs_node.name
                    if obs_name and obs_name != var_name:
                        input_map[var_name] = input_map[var_name].difference({obs_name})
                        input_map[obs_name] = input_map[obs_name].union({var_name})
                        break
                    elif (
                        # for cases where observations are cast to a certain dtype
                        # see issue 5795: https://github.com/pymc-devs/pymc/issues/5795
                        obs_node.owner
                        and isinstance(obs_node.owner.op, Elemwise)
                        and isinstance(obs_node.owner.op.scalar_op, Cast)
                    ):
                        # we can retrieve the observation node by going up the graph
                        obs_node = obs_node.owner.inputs[0]
                    else:
                        break

        return input_map

    def get_plates(
        self,
        var_names: Iterable[VarName] | None = None,
    ) -> list[Plate]:
        """Rough but surprisingly accurate plate detection.

        Just groups by the shape of the underlying distribution.  Will be wrong
        if there are two plates with the same shape.

        Returns
        -------
        dict
            Maps plate labels to the set of ``VarName``s inside the plate.
        """
        plates = defaultdict(set)

        # TODO: Evaluate all RV shapes at once
        #       This should help find discrepencies, and
        #       avoids unnecessary function compiles for determining labels.
        dim_lengths: dict[str, int] = {
            dim_name: fast_eval(value).item() for dim_name, value in self.model.dim_lengths.items()
        }
        var_shapes: dict[str, tuple[int, ...]] = {
            var_name: tuple(fast_eval(self.model[var_name].shape))
            for var_name in self.vars_to_plot(var_names)
        }

        for var_name in self.vars_to_plot(var_names):
            shape: tuple[int, ...] = var_shapes[var_name]
            if var_name in self.model.named_vars_to_dims:
                # The RV is associated with `dims` information.
                names = []
                lengths = []
                for d, dname in enumerate(self.model.named_vars_to_dims[var_name]):
                    names.append(dname)
                    lengths.append(dim_lengths.get(dname, shape[d]))

                dim_info = DimInfo(
                    names=tuple(names),
                    lengths=tuple(lengths),
                )
            else:
                # The RV has no `dims` information.
                dim_size = len(shape)
                dim_info = DimInfo(
                    names=tuple([None] * dim_size),
                    lengths=tuple(shape),
                )

            v = self.model[var_name]
            node_type = get_node_type(var_name, self.model)
            var = NodeInfo(var=v, node_type=node_type)
            plates[dim_info].add(var)

        return [
            Plate(
                dim_info=dim_info,
                variables=list(variables),
            )
            for dim_info, variables in plates.items()
        ]

    def edges(
        self,
        var_names: Iterable[VarName] | None = None,
    ) -> list[tuple[VarName, VarName]]:
        """Get edges between the variables in the model.

        Parameters
        ----------
        var_names : iterable of str, optional
            Subset of variables to be plotted that identify a subgraph with respect to the entire model graph

        Returns
        -------
        list of tuple
            List of edges between the variables in the model.

        """
        return [
            (VarName(child.replace(":", "&")), VarName(parent.replace(":", "&")))
            for child, parents in self.make_compute_graph(var_names=var_names).items()
            for parent in parents
        ]


def make_graph(
    name: str,
    plates: list[Plate],
    edges: list[tuple[VarName, VarName]],
    formatting: str = "plain",
    save=None,
    figsize=None,
    dpi=300,
    node_formatters: NodeTypeFormatterMapping | None = None,
    create_plate_label: PlateLabelFunc = create_plate_label_with_dim_length,
):
    """Make graphviz Digraph of PyMC model

    Returns
    -------
    graphviz.Digraph
    """
    try:
        import graphviz
    except ImportError:
        raise ImportError(
            "This function requires the python library graphviz, along with binaries. "
            "The easiest way to install all of this is by running\n\n"
            "\tconda install -c conda-forge python-graphviz"
        )

    node_formatters = node_formatters or {}
    node_formatters = update_node_formatters(node_formatters)

    graph = graphviz.Digraph(name)
    for plate in plates:
        if plate.dim_info:
            # must be preceded by 'cluster' to get a box around it
            plate_label = create_plate_label(plate.dim_info)
            plate_name = f"cluster{plate_label}"

            with graph.subgraph(name=plate_name) as sub:
                for var in plate.variables:
                    _make_node(
                        node=var,
                        formatting=formatting,
                        node_formatters=node_formatters,
                        add_node=sub.node,
                    )
                # plate label goes bottom right
                sub.attr(label=plate_label, labeljust="r", labelloc="b", style="rounded")
        else:
            for var in plate.variables:
                _make_node(
                    node=var,
                    formatting=formatting,
                    node_formatters=node_formatters,
                    add_node=graph.node,
                )

    for child, parent in edges:
        graph.edge(parent, child)

    if save is not None:
        width, height = (None, None) if figsize is None else figsize
        base, ext = path.splitext(save)
        if ext:
            ext = ext.replace(".", "")
        else:
            ext = "png"
        graph_c = graph.copy()
        graph_c.graph_attr.update(size=f"{width},{height}!")
        graph_c.graph_attr.update(dpi=str(dpi))
        graph_c.render(filename=base, format=ext, cleanup=True)

    return graph


def make_networkx(
    name: str,
    plates: list[Plate],
    edges: list[tuple[VarName, VarName]],
    formatting: str = "plain",
    node_formatters: NodeTypeFormatterMapping | None = None,
    create_plate_label: PlateLabelFunc = create_plate_label_with_dim_length,
):
    """Make networkx Digraph of PyMC model

    Returns
    -------
    networkx.Digraph
    """
    try:
        import networkx
    except ImportError:
        raise ImportError(
            "This function requires the python library networkx, along with binaries. "
            "The easiest way to install all of this is by running\n\n"
            "\tconda install networkx"
        )

    node_formatters = node_formatters or {}
    node_formatters = update_node_formatters(node_formatters)

    graphnetwork = networkx.DiGraph(name=name)
    for plate in plates:
        if plate.dim_info:
            # # must be preceded by 'cluster' to get a box around it

            plate_label = create_plate_label(plate.dim_info)
            plate_name = f"cluster{plate_label}"
            subgraphnetwork = networkx.DiGraph(name=plate_name, label=plate_label)

            for var in plate.variables:
                _make_node(
                    node=var,
                    node_formatters=node_formatters,
                    cluster=plate_name,
                    formatting=formatting,
                    add_node=subgraphnetwork.add_node,
                )
            for sgn in subgraphnetwork.nodes:
                networkx.set_node_attributes(
                    subgraphnetwork,
                    {sgn: {"labeljust": "r", "labelloc": "b", "style": "rounded"}},
                )
            node_data = {
                e[0]: e[1] for e in graphnetwork.nodes(data=True) & subgraphnetwork.nodes(data=True)
            }

            graphnetwork = networkx.compose(graphnetwork, subgraphnetwork)
            networkx.set_node_attributes(graphnetwork, node_data)
            graphnetwork.graph["name"] = name
        else:
            for var in plate.variables:
                _make_node(
                    node=var,
                    formatting=formatting,
                    node_formatters=node_formatters,
                    add_node=graphnetwork.add_node,
                )

    for child, parents in edges:
        graphnetwork.add_edge(parents, child)

    return graphnetwork


[docs] def model_to_networkx( model=None, *, var_names: Iterable[VarName] | None = None, formatting: str = "plain", node_formatters: NodeTypeFormatterMapping | None = None, include_dim_lengths: bool = True, ): """Produce a networkx Digraph from a PyMC model. Requires networkx, which may be installed most easily with:: conda install networkx Alternatively, you may install using pip with:: pip install networkx See https://networkx.org/documentation/stable/ for more information. Parameters ---------- model : Model The model to plot. Not required when called from inside a modelcontext. var_names : iterable of str, optional Subset of variables to be plotted that identify a subgraph with respect to the entire model graph formatting : str, optional one of { "plain" } node_formatters : dict, optional A dictionary mapping node types to functions that return a dictionary of node attributes. Check out the networkx documentation for more information how attributes are added to nodes: https://networkx.org/documentation/stable/reference/classes/generated/networkx.Graph.add_node.html include_dim_lengths : bool Include the dim length in the plate label. Default is True. Examples -------- How to plot the graph of the model. .. code-block:: python import numpy as np from pymc import HalfCauchy, Model, Normal, model_to_networkx J = 8 y = np.array([28, 8, -3, 7, -1, 1, 18, 12]) sigma = np.array([15, 10, 16, 11, 9, 11, 10, 18]) with Model() as schools: eta = Normal("eta", 0, 1, shape=J) mu = Normal("mu", 0, sigma=1e6) tau = HalfCauchy("tau", 25) theta = mu + tau * eta obs = Normal("obs", theta, sigma=sigma, observed=y) model_to_networkx(schools) Add custom attributes to Free Random Variables and Observed Random Variables nodes. .. code-block:: python node_formatters = { "Free Random Variable": lambda var: {"shape": "circle", "label": var.name}, "Observed Random Variable": lambda var: {"shape": "square", "label": var.name}, } model_to_networkx(schools, node_formatters=node_formatters) """ if "plain" not in formatting: raise ValueError(f"Unsupported formatting for graph nodes: '{formatting}'. See docstring.") if formatting != "plain": warnings.warn( "Formattings other than 'plain' are currently not supported.", UserWarning, stacklevel=2, ) model = pm.modelcontext(model) graph = ModelGraph(model) return make_networkx( name=model.name, plates=graph.get_plates(var_names=var_names), edges=graph.edges(var_names=var_names), formatting=formatting, node_formatters=node_formatters, create_plate_label=create_plate_label_with_dim_length if include_dim_lengths else create_plate_label_without_dim_length, )
[docs] def model_to_graphviz( model=None, *, var_names: Iterable[VarName] | None = None, formatting: str = "plain", save: str | None = None, figsize: tuple[int, int] | None = None, dpi: int = 300, node_formatters: NodeTypeFormatterMapping | None = None, include_dim_lengths: bool = True, ): """Produce a graphviz Digraph from a PyMC model. Requires graphviz, which may be installed most easily with conda install -c conda-forge python-graphviz Alternatively, you may install the `graphviz` binaries yourself, and then `pip install graphviz` to get the python bindings. See http://graphviz.readthedocs.io/en/stable/manual.html for more information. Parameters ---------- model : pm.Model The model to plot. Not required when called from inside a modelcontext. var_names : iterable of variable names, optional Subset of variables to be plotted that identify a subgraph with respect to the entire model graph formatting : str, optional one of { "plain" } save : str, optional If provided, an image of the graph will be saved to this location. The format is inferred from the file extension. figsize : tuple[int, int], optional Width and height of the figure in inches. If not provided, uses the default figure size. It only affect the size of the saved figure. dpi : int, optional Dots per inch. It only affects the resolution of the saved figure. The default is 300. node_formatters : dict, optional A dictionary mapping node types to functions that return a dictionary of node attributes. Check out graphviz documentation for more information on available attributes. https://graphviz.org/docs/nodes/ include_dim_lengths : bool Include the dim lengths in the plate label. Default is True. Examples -------- How to plot the graph of the model. .. code-block:: python import numpy as np from pymc import HalfCauchy, Model, Normal, model_to_graphviz J = 8 y = np.array([28, 8, -3, 7, -1, 1, 18, 12]) sigma = np.array([15, 10, 16, 11, 9, 11, 10, 18]) with Model() as schools: eta = Normal("eta", 0, 1, shape=J) mu = Normal("mu", 0, sigma=1e6) tau = HalfCauchy("tau", 25) theta = mu + tau * eta obs = Normal("obs", theta, sigma=sigma, observed=y) model_to_graphviz(schools) Note that this code automatically plots the graph if executed in a Jupyter notebook. If executed non-interactively, such as in a script or python console, the graph needs to be rendered explicitly: .. code-block:: python # creates the file `schools.pdf` model_to_graphviz(schools).render("schools") Display Free Random Variables and Observed Random Variables nodes with custom formatting. .. code-block:: python node_formatters = { "Free Random Variable": lambda var: {"shape": "circle", "label": var.name}, "Observed Random Variable": lambda var: {"shape": "square", "label": var.name}, } model_to_graphviz(schools, node_formatters=node_formatters) """ if "plain" not in formatting: raise ValueError(f"Unsupported formatting for graph nodes: '{formatting}'. See docstring.") if formatting != "plain": warnings.warn( "Formattings other than 'plain' are currently not supported.", UserWarning, stacklevel=2, ) model = pm.modelcontext(model) graph = ModelGraph(model) return make_graph( model.name, plates=graph.get_plates(var_names=var_names), edges=graph.edges(var_names=var_names), formatting=formatting, save=save, figsize=figsize, dpi=dpi, node_formatters=node_formatters, create_plate_label=create_plate_label_with_dim_length if include_dim_lengths else create_plate_label_without_dim_length, )