Errors on usage.

[daisylab]

Hi. I'm a graduate student from korea, majoring computer science.

I think LNN is a great project what I'm looking for. I really appreciate it. Thanks.

However, I found some errors in the usage page.

https://ibm.github.io/LNN/education/examples/reasoning.html#more-complex-reasoning-example

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

model = Model()  # Instantiate a model.
x, y, z, w = map(Variable, ['x', 'y', 'z', 'w'])

# Define and add predicates to the model.
owns = model['owns'] = Predicate('owns', 2)  # binary predicate
missile = model['missile'] = Predicate('missile')
american = model['american'] = Predicate('american')
enemy = model['enemy'] = Predicate('enemy', 2)
hostile = model['hostile'] = Predicate('hostile')
criminal = model['criminal'] = Predicate('criminal')
weapon = model['weapon'] = Predicate('weapon')
sells = model['sells'] = Predicate('sells', 3)  # ternary predicate

which tells that

      5 x, y, z, w = map(Variable, ['x', 'y', 'z', 'w'])
      7 # Define and add predicates to the model.
----> 8 owns = model['owns'] = Predicate('owns', 2)  # binary predicate
      9 missile = model['missile'] = Predicate('missile')
     10 american = model['american'] = Predicate('american')

TypeError: 'Model' object does not support item assignment

I'm new to LNN, and not yet fully understand it. Can you please tell me how to fix this error? Thank you very much.

[daisylab]

A few comments.

So i deleted the item assignment from the code.

# Define and add predicates to the model.
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

It seems to work, however, I got another error message from that code.

# Define and add the background knowledge to  the model.
america_enemies = (
    ForAll(x, Implies(enemy(x, (y, 'America')), 
                      hostile(x),
                      join=Join.OUTER),
           join=Join.OUTER, 
           world=World.AXIOM)
    )
model.add_knowledge(america_enemies)

which gave me an error message:

File ~/home/lnn-homework/complex.py:19
     15 sells = Predicate('sells', 3)  # ternary predicate
     17 # Define and add the background knowledge to  the model.
     18 america_enemies = (
---> 19     ForAll(x, Implies(enemy(x, (y, 'America')), 
     20                       hostile(x),
     21                       join=Join.OUTER),
     22            join=Join.OUTER, 
     23            world=World.AXIOM)
     24     )
     25 model.add_knowledge(america_enemies)

File ~/.virtualenvs/lnn/lib/python3.10/site-packages/lnn/symbolic/logic/leaf_formula.py:92, in Predicate.__call__(self, *args, **kwds)
     75 def __call__(self, *args, **kwds):
     76     r"""A called first-order logic predicate
     77 
     78     This correctly instantiates a predicate with variables - which is required when
   (...)
     90     an ordered unique collection (list).
     91     """
---> 92     return super().__call__(*args, **kwds)

File ~/.virtualenvs/lnn/lib/python3.10/site-packages/lnn/symbolic/logic/formula.py:689, in Formula.__call__(self, bind, *variables)
    687     variable_objects.append(v)
    688 else:
--> 689     raise TypeError(f"expected variable, received {type(v)}")
    691 if v in bind:
    692     if isinstance(bind[v], str):  # a single binding

TypeError: expected variable, received <class 'tuple'>

In [3]: 

[KyleErwin]

Hi @daisylab. Nice to meet you. It is great to hear that you are using the LNN.

Thank you for pointing this out. I will try to replicate the issue. In the meantime, please see here for a working example of the same code. Let me know if you have any questions.

[daisylab]

Thank you for the working example. It runs smoothly without errors. It greatly helped me.

After testing the code you mentioned above, I myself tried to modify the example code, and I got the right result, a.k.a, {'West'}.

Below is my modification to the example code:

# The code snippet below shows the declaration of the model and all the
# predicates.

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

model = Model()  # Instantiate a model.
x, y, z, w = map(Variable, ['x', 'y', 'z', 'w'])

# Define and add predicates to the model.
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

# The code snippet below shows an example of adding the background
# knowledge into the model. All the other rules can be added to the
# model similarly.

model.add_knowledge(
    ForAll(x, y,
           Implies(enemy(x, y, bind={y: 'America'}),
                   hostile(x))),
    ForAll(x, y, z,
           Implies(And(american(x), weapon(y), sells(x, y, z),
                       hostile(z)),
                   criminal(x))),
    ForAll(x, y, z,
           Implies(And(missile(x), owns(y, x, bind={y: 'Nono'})),
                   sells(z, x, y, bind={z: 'West', y: 'Nono'}))),
    ForAll(x,
           Implies(missile(x), weapon(x)))

)

# The code snippet below shows how to add the query into the model.

query = Exists(x, criminal(x))

# model.add_knowledge(query)
model.set_query(query)

# Finally, we can add facts into the model.

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

# Then we are ready to perform inference and answer queries (aka
# theorem proving):

model.infer()

print(model[query].true_groundings)

Thank you for your advice.

[NaweedAghmad]

The above example API is outdated and covered in the unittest. I don't see a question or issue, so closing it off.