neurolang.probabilistic.weighted_model_counting module

Implementation of the weighted model counting approach through knowledge compilation.

class neurolang.probabilistic.weighted_model_counting.DeterministicFactSetTranslation(*args, **kwargs)

Bases: PatternWalker

Attributes:

patterns

Property holding an iterator of triplets (pattern, guard, action).

Methods

match(expression)	Find the action for a given expression by going through the patterns.
pattern_match(pattern, expression)	Return True if pattern matches expression.
type

deterministic_fact_set_constant
pattern_match_expression
pattern_match_expression_parameters
pattern_match_expression_tuple
pattern_match_tuple
walk

deterministic_fact_set_constant(deterministic_set)

type = typing.Any

class neurolang.probabilistic.weighted_model_counting.EliminateSuperfluousProjectionMixin(*args, **kwargs)

Bases: PatternWalker

Attributes:

patterns

Property holding an iterator of triplets (pattern, guard, action).

Methods

match(expression)	Find the action for a given expression by going through the patterns.
pattern_match(pattern, expression)	Return True if pattern matches expression.
type

eliminate_superfluous_projection
pattern_match_expression
pattern_match_expression_parameters
pattern_match_expression_tuple
pattern_match_tuple
walk

eliminate_superfluous_projection(expression)

type = typing.Any

class neurolang.probabilistic.weighted_model_counting.RAQueryOptimiser(*args, **kwargs)

Bases: PushInSelections, ExpressionWalker

Attributes:

patterns

Property holding an iterator of triplets (pattern, guard, action).

Methods

match(expression)	Find the action for a given expression by going through the patterns.
pattern_match(pattern, expression)	Return True if pattern matches expression.
type

pattern_match_expression
pattern_match_expression_parameters
pattern_match_expression_tuple
pattern_match_tuple
process_expression
process_iterable_argument
push_and_rename_in_naturaljoin
push_selection_in_extended_projection
push_selection_in_groupby
push_selection_in_left
push_selection_in_leftnaturaljoin_left
push_selection_in_leftnaturaljoin_right
push_selection_in_outernaturaljoin_left
push_selection_in_outernaturaljoin_right
push_selection_in_projection
push_selection_in_replace_null
push_selection_in_right
walk

type = typing.Any

class neurolang.probabilistic.weighted_model_counting.SDDWMCSemiRingSolver(*args, **kwargs)

Bases: EliminateSuperfluousProjectionMixin, RenameOptimizations, DeterministicFactSetTranslation, RelationalAlgebraProvenanceExpressionSemringSolver

Attributes:

patterns

Property holding an iterator of triplets (pattern, guard, action).

Methods

match(expression)	Find the action for a given expression by going through the patterns.
pattern_match(pattern, expression)	Return True if pattern matches expression.
type

aggregate
arithmetic_string_expression
concatenate_constant_column
convert_rename_column
cycle_in_build_provenance_algebra_set
deterministic_fact_set
deterministic_fact_set_constant
difference
eliminate_superfluous_projection
eliminate_trivial_number_columns
eliminate_trivial_projection_number_columns
eliminate_trivial_selection_number_columns
extended_projection
generate_sdd_expression
get_new_bernoulli_variable
merge_nested_rename_columns
merge_rename_column_group_by
merge_rename_columns_extended_projection
name_columns_int_rap
name_columns_rap
number_columns_rap
pattern_match_expression
pattern_match_expression_parameters
pattern_match_expression_tuple
pattern_match_tuple
probabilistic_choice_set
probabilistic_fact_set
probabilistic_fact_set_invalid
process_expression
process_iterable_argument
projection_rap
projection_rap_columnint
projection_rap_columnint_subset
prov_arithmetic_operation
prov_concatenate_constant_column
prov_extended_projection
prov_naturaljoin
prov_product
prov_rename_column
prov_rename_columns
prov_union
push_rename_past_groupby
push_rename_past_projection
push_rename_past_selection
ra_constant
ra_difference
ra_equijoin
ra_full_outer_naturaljoin
ra_intersection
ra_left_naturaljoin
ra_name_columns
ra_naturaljoin
ra_number_columns
ra_product
ra_projection
ra_rename_column
ra_rename_columns
ra_symbol
ra_union
remove_trivial_rename
replace_null
selection_between_columns
selection_by_constant
selection_general_selection_by_constant
selection_provenance_set
selection_rap_eq_columnint
selection_rap_eq_columnint_columnint
set_destroy
simplify_renames_name
split_rename_left_naturaljoin
split_rename_naturaljoin
split_rename_outer_naturaljoin
switch_rename_replace_null
walk
wmc_weights

deterministic_fact_set(deterministic_set)

generate_sdd_expression(adjusted_probability, tag_expressions, previous_expression)

get_new_bernoulli_variable(probability)

probabilistic_choice_set(prob_fact_set)

probabilistic_fact_set(prob_fact_set)

probabilistic_fact_set_invalid(prob_fact_set)

type = typing.Any

wmc_weights()

class neurolang.probabilistic.weighted_model_counting.SemiRingRAPToSDD(var_count, symbols_to_literals=None)

Bases: PatternWalker

Attributes:

patterns

Property holding an iterator of triplets (pattern, guard, action).

Methods

match(expression)	Find the action for a given expression by going through the patterns.
pattern_match(pattern, expression)	Return True if pattern matches expression.
type

add
eq
expression_block
mul
neg
or_
pattern_match_expression
pattern_match_expression_parameters
pattern_match_expression_tuple
pattern_match_tuple
symbol
walk

add(expression)

eq(expression)

expression_block(expression)

mul(expression)

neg(expression)

or_(expression)

symbol(expression)

type = typing.Any

class neurolang.probabilistic.weighted_model_counting.WMCSemiRingSolver(*args, **kwargs)

Bases: EliminateSuperfluousProjectionMixin, DeterministicFactSetTranslation, RelationalAlgebraProvenanceExpressionSemringSolver

Attributes:

patterns

Property holding an iterator of triplets (pattern, guard, action).

Methods

match(expression)	Find the action for a given expression by going through the patterns.
pattern_match(pattern, expression)	Return True if pattern matches expression.
type

aggregate
arithmetic_string_expression
concatenate_constant_column
cycle_in_build_provenance_algebra_set
deterministic_fact_set
deterministic_fact_set_constant
difference
eliminate_superfluous_projection
extended_projection
name_columns_int_rap
name_columns_rap
number_columns_rap
pattern_match_expression
pattern_match_expression_parameters
pattern_match_expression_tuple
pattern_match_tuple
probabilistic_choice_set
probabilistic_fact_set
probabilistic_fact_set_invalid
process_expression
process_iterable_argument
projection_rap
projection_rap_columnint
projection_rap_columnint_subset
prov_arithmetic_operation
prov_concatenate_constant_column
prov_extended_projection
prov_naturaljoin
prov_product
prov_rename_column
prov_rename_columns
prov_union
ra_constant
ra_difference
ra_equijoin
ra_full_outer_naturaljoin
ra_intersection
ra_left_naturaljoin
ra_name_columns
ra_naturaljoin
ra_number_columns
ra_product
ra_projection
ra_rename_column
ra_rename_columns
ra_symbol
ra_union
replace_null
selection_between_columns
selection_by_constant
selection_general_selection_by_constant
selection_provenance_set
selection_rap_eq_columnint
selection_rap_eq_columnint_columnint
set_destroy
walk

deterministic_fact_set(deterministic_set)

probabilistic_choice_set(prob_choice_set)

probabilistic_fact_set(prob_fact_set)

probabilistic_fact_set_invalid(prob_fact_set)

type = typing.Any

neurolang.probabilistic.weighted_model_counting.build_global_sdd_model_rows(solver, literal_probabilities)

neurolang.probabilistic.weighted_model_counting.generate_probability_table(solver)

neurolang.probabilistic.weighted_model_counting.generate_weights(symbol_probs, literals_to_symbols, extras=0)

neurolang.probabilistic.weighted_model_counting.model_count_and_per_row_probability(model, solver, initial_var_count, n)

neurolang.probabilistic.weighted_model_counting.perform_wmc(solver, sdd_compiler, sdd_program, prob_set_program, prob_set_result)

neurolang.probabilistic.weighted_model_counting.prepare_initial_query(query_predicate)

neurolang.probabilistic.weighted_model_counting.sdd_compilation(prob_set_result)

neurolang.probabilistic.weighted_model_counting.sdd_compilation_and_wmc(prob_set_result, solver)

neurolang.probabilistic.weighted_model_counting.sdd_solver_global_model(solver, set_probabilities)

neurolang.probabilistic.weighted_model_counting.sdd_solver_per_individual_row(solver, set_probabilities)

neurolang.probabilistic.weighted_model_counting.solve_marg_query(rule, cpl)

Solve a MARG query on a CP-Logic program.

Parameters:

queryImplication

Consequent must be of type Condition. MARG query of the form ans(x) :- P(x).

cpl_programCPLogicProgram

CP-Logic program on which the query should be solved.

Returns:

ProvenanceAlgebraSet

Provenance set labelled with probabilities for each tuple in the result set.

neurolang.probabilistic.weighted_model_counting.solve_succ_query(query_predicate, cpl_program, per_row_model=True, run_relational_algebra_solver=True)

Obtain the solution of a SUCC query on a CP-Logic program.

The SUCC query must take the form

SUCC[ P(x) ]

neurolang.probabilistic.weighted_model_counting.solve_succ_query_boolean_diagram(query_predicate, cpl_program, run_relational_algebra_solver=False)

Obtain the solution of a SUCC query on a CP-Logic program.

The SUCC query must take the form

SUCC[ P(x) ]

neurolang.probabilistic.weighted_model_counting.solve_succ_query_sdd_direct(query_predicate, cpl_program, per_row_model=True, run_relational_algebra_solver=True)

Obtain the solution of a SUCC query on a CP-Logic program.

The SUCC query must take the form

SUCC[ P(x) ]