example error

[PolKul]

When trying to run your example I get the following error: TypeError: expected variable, received <class 'tuple'>

american_enemies = ( ForAll(x, Implies(enemy(x, (y,'America')), hostile(x), join=Join.OUTER), join=Join.OUTER, world=World.AXIOM) )

When I replace Join.OUTER with Join.INNER it compiles well, but still doesn't give me the correct result. What is the meaning of these Join constants? And what can be wrong with the formula?

[PolKul]

Let me provide the full code as I made it based on your example. It gives me an error when I try to run it.

from lnn import (Predicate, Variable, And, Join,
                 Exists, Implies, ForAll, Model, Fact, World)

model = Model()

# Variablle
x, y, z, w = map(Variable, ['x', 'y', 'z', 'w'])

# Define and add predicates to the model.
owns = Predicate(arity=2, name='owns')
missile = Predicate(arity=1, name='missile')
american = Predicate(arity=1, name='american')
enemy = Predicate(arity=2, name='enemy')
hostile = Predicate(arity=1, name='hostile')
criminal = Predicate(arity=1, name='criminal')
weapon = Predicate(arity=1, name='weapon')
sells = Predicate(arity=3, name='sells')

# Define and add the background knowledge to  the model.
is_criminal = ForAll(x, y, z,
                     Implies(
                         And(american(x), weapon(y), sells(x, y, z), hostile(z)), criminal(x),
                         join=Join.OUTER),
                     name='is-criminal', join=Join.OUTER, world=World.AXIOM)

is_selling = ForAll(x,
                    Implies(
                        And(missile(x), owns('Nono', x)), sells('West', x, 'Nono'),
                        name='is-selling', join=Join.OUTER),
                    name='is-selling', join=Join.OUTER, world=World.AXIOM)

missile_is_weapon = ForAll(x,
                           Implies(missile(x), weapon(x),
                           name='is-weapon', join=Join.OUTER),
                    name='missile-is-weapon', join=Join.OUTER, world=World.AXIOM)

american_enemies = (
    ForAll(x, Implies(enemy(x, "America"),
                      hostile(x),
                      join=Join.OUTER),
           join=Join.OUTER,
           world=World.AXIOM)
)

# Query
query = Exists(x, criminal(x), name='criminal-west')

# Add predicates and rules to the model
model.add_knowledge(owns, missile, american, enemy, hostile, criminal, weapon, sells, is_criminal, is_selling, missile_is_weapon, american_enemies, query)

# Add facts to the model
model.add_data({
    owns: {('Nono', 'M1'): Fact.TRUE},
    missile: {'M1': Fact.TRUE},
    american: {'West': Fact.TRUE},
    enemy: {('Nono', 'America'): Fact.TRUE},
})

# Perform inference
steps, facts_inferred = model.infer()
print(query.true_groundings)

The error is as follows:

Traceback (most recent call last):
  File "/lnn_test.py", line 60, in <module>
    steps, facts_inferred = model.infer()
  File "/venv/lib/python3.10/site-packages/lnn/model.py", line 497, in infer
    return self._infer(
  File "/venv/lib/python3.10/site-packages/lnn/model.py", line 541, in _infer
    bounds_diff += self._traverse_execute(
  File "/venv/lib/python3.10/site-packages/lnn/model.py", line 438, in _traverse_execute
    val = getattr(node, func)(**kwds) if hasattr(node, func) else None
  File "/venv/lib/python3.10/site-packages/lnn/symbolic/logic/connective_neuron.py", line 93, in upward
    upward_bounds = _gm.upward_bounds(self, self.operands, groundings)
  File "/venv/lib/python3.10/site-packages/lnn/symbolic/_gm.py", line 39, in upward_bounds
    result = _operational_bounds(self, Direction.UPWARD, operands, groundings)
  File "/venv/lib/python3.10/site-packages/lnn/symbolic/_gm.py", line 166, in _operational_bounds
    ground_tuples, ground_objects = _hash_join_outer(self, tmp_bindings)
  File "/venv/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/venv/lib/python3.10/site-packages/lnn/symbolic/_gm.py", line 437, in _hash_join_outer
    g_obj = operands[i]._ground(op_tup)
  File "/venv/lib/python3.10/site-packages/lnn/symbolic/logic/formula.py", line 905, in _ground
    raise Exception(
Exception: expected grounding length to be of arity 3, received ('M1',)

[riccardopinosio]

I get the same/very similar error. My code:

from lnn import (Predicate, Variable, Join, And, Exists, Implies, ForAll, Model,
Fact, World)

model = Model()

x, y, z, w = map(Variable, ["x", "y", "z", "w"])

owns = Predicate('owns', 2)  # binary predicate
missile = Predicate('missile')
american = Predicate('american')
enemy = Predicate('enemy', 2)
hostile = Predicate('hostile')
criminal = Predicate('criminal')
weapon = Predicate('weapon')
sells = Predicate('sells', 3)  # ternary predicate

america_enemies = (
    ForAll(x, 
    Implies(enemy(x, 'America'), hostile(x), join=Join.OUTER),
    join=Join.OUTER,
    world=World.AXIOM
    )
)

model.add_knowledge(america_enemies)
model.add_knowledge(*[
    ForAll(x, Implies(missile(x), weapon(x))),
    ForAll(x, Implies(
        And(missile(x), owns('nono', x)), sells('west', x, 'nono')
        )
        ),
    ForAll(x, y, z, Implies(
        And(american(x), weapon(y), sells(x,y,z), hostile(z)),
        criminal(x)
    ))
])

model.add_data({
    owns: {
        ('nono', 'm1'): Fact.TRUE
    },
    missile: {
        "m1": Fact.TRUE,
        "m2": Fact.TRUE,
        "m3": Fact.TRUE
    },
    american: {
        "west": Fact.TRUE
    },
    enemy: {
        ('nono', 'america'): Fact.TRUE
    }
}
)

query = Exists(x, criminal(x))
model.add_knowledge(query)
model.infer()

error:

Exception                                 Traceback (most recent call last)
Cell In [16], line 1
----> 1 model.infer()

File ~/repositories/neurosymb/LNN/lnn/model.py:497, in Model.infer(self, direction, source, max_steps, lifted, **kwds)
    494 if lifted:
    495     self.lift(lifted)
--> 497 return self._infer(
    498     direction=direction,
    499     source=source,
    500     max_steps=max_steps,
    501     lifted=False,
    502     **kwds,
    503 )

File ~/repositories/neurosymb/LNN/lnn/model.py:541, in Model._infer(self, direction, source, max_steps, **kwds)
    539 bounds_diff = 0.0
    540 for d in direction:
--> 541     bounds_diff += self._traverse_execute(
    542         d.value.lower(), d, source, **kwds
    543     )
    544 converged_bounds = (
    545     True
    546     if direction in ([[Direction.UPWARD], [Direction.DOWNWARD]])
...
    908         )
    909 else:
    910     if self.num_unique_vars != 1 and arity_match:

Exception: expected grounding length to be of arity 2, received ('nono',)