Transforming Array Expression Graphs

class pytato.transform.CopyMapper

Performs a deep copy of a pytato.array.Array. The typical use of this mapper is to override individual map_ methods in subclasses to permit term rewriting on an expression graph.

class pytato.transform.DependencyMapper

Maps a pytato.array.Array to a frozenset of pytato.array.Array’s it depends on. .. warning:

This returns every node in the graph! Consider a custom
:class:`CombineMapper` or a :class:`SubsetDependencyMapper` instead.

class pytato.transform.InputGatherer

Mapper to combine all instances of pytato.array.InputArgumentBase that an array expression depends on.

class pytato.transform.SizeParamGatherer

Mapper to combine all instances of pytato.array.SizeParam that an array expression depends on.

class pytato.transform.SubsetDependencyMapper(universe)

Mapper to combine the dependencies of an expression that are a subset of universe.

class pytato.transform.WalkMapper

A mapper that walks over all the arrays in a pytato.Array.

Users may override the specific mapper methods in a derived class or override WalkMapper.visit() and WalkMapper.post_visit().

visit(expr)

If this method returns True, expr is traversed during the walk. If this method returns False, expr is not traversed as a part of the walk.

Return type

bool

post_visit(expr)

Callback after expr has been traversed.

Return type

None

class pytato.transform.CachedWalkMapper

WalkMapper that visits each node in the DAG exactly once. This loses some information compared to WalkMapper as a node is visited only from one of its predecessors.

class pytato.transform.TopoSortMapper

A mapper that creates a list of nodes in topological order.

pytato.transform.copy_dict_of_named_arrays(source_dict, copy_mapper)

Copy the elements of a DictOfNamedArrays into a DictOfNamedArrays.

Parameters

• source_dict (DictOfNamedArrays) – The DictOfNamedArrays to copy

• copy_mapper (CopyMapper) – A mapper that performs copies different array types

Return type

DictOfNamedArrays

Returns

A new DictOfNamedArrays containing copies of the items in source_dict

pytato.transform.get_dependencies(expr)

Returns the dependencies of each named array in expr.

Return type

Dict[str, FrozenSet[Array]]

Visualizing Array Expression Graphs

pytato.get_dot_graph(result)

Return a string in the dot language depicting the graph of the computation of result.

Parameters

result (Union[Array, DictOfNamedArrays]) – Outputs of the computation (cf. pytato.generate_loopy()).

Return type

str

pytato.show_dot_graph(result)

Show a graph representing the computation of result in a browser.

Parameters

result (Union[Array, DictOfNamedArrays]) – Outputs of the computation (cf. pytato.generate_loopy()).

Return type

None

Utilities and Diagnostics

Helper routines

pytato.utils.are_shape_components_equal(dim1, dim2)

Returns True iff dim1 and dim2 are have equal SizeParam coefficients in their expressions.

Return type

bool

pytato.utils.are_shapes_equal(shape1, shape2)

Returns True iff shape1 and shape2 have the same dimensionality and the correpsonding components are equal as defined by are_shape_components_equal().

Return type

bool

pytato.utils.get_shape_after_broadcasting(exprs)

Returns the shape after broadcasting exprs in an operation.

Return type

Tuple[Union[int, Array], …]

pytato.utils.dim_to_index_lambda_components(expr, vng=None)

Returns the scalar expressions and bindings to use the shape component within an index lambda.

>>> n = pt.make_size_param("n")
>>> expr, bnds = dim_to_index_lambda_components(3*n+8, UniqueNameGenerator())
>>> print(expr)
3*_in + 8
>>> bnds  
{'_in': <pytato.array.SizeParam ...>}

Return type

Tuple[Union[Number, bool_, bool, Expression], Dict[str, SizeParam]]

Pytato-specific exceptions

class pytato.diagnostic.NameClashError

Raised when 2 non-identical InputArgumentBase’s are reachable in an Array’s DAG and share the same name. Here, we refer to 2 objects a and b as being identical iff a is b.

class pytato.diagnostic.CannotBroadcastError