Decoder

Making Decisions

Making choices and decisions is a core part of any intelligence system, and we wanted to make that as easy as possible. The Decoder can be used to make decisions or navigate choices with minimal configuration or code.

Basic Decoder

Here is a simple example of how to create a Decoder that returns the family of an animal. Decoders always return a list of values, and this can be controlled by the choice_count argument which defaults to 1.

from dandy import Decoder

class AnimalFamilyDecoder(Decoder):
    mapping_keys_description = 'Animal Sounds'
    mapping = {
        'barking': 'dog',
        'meowing': 'cat',
        'quacking': 'duck'
    }

animal_families = AnimalFamilyDecoder().process(
    'I was out on a walk and heard some barking', 
    max_return_values=1
)

print(animal_families[0])

dog

Advanced Decoder

Decoders can also be nested indefinitely and link to any value you want. While a decoder is traversing if it comes across another Decoder it will continue to traverse to get the values nested within.

from dandy import Decoder

class MathBook:
    def __str__(self):
        return 'Math Book'

golf = 'Golf'

class LearningDecoder(Decoder):
    mapping_keys_description = 'Learning Subjects'
    mapping = {
        'history': 'Social Studies Book',
        'science': 'Laboratory Book',
        'numbers': MathBook()
    }

class ActivityDecoder(Decoder):
    mapping_keys_description = 'Physical Activities'
    mapping = {
        'running around outdoors': Decoder(
            mapping_keys_description='Activities',
            mapping={
            'throwing': 'Ball',
            'flinging': 'Frisbee',
            'walking': golf
        }),
        'playing with friends': Decoder(
            mapping_keys_description='Activities',
            mapping={
            'kicking': 'soccer',
            'swinging': 'baseball',
            'climbing': 'jungle gym'    
        }),
        'being inside': Decoder(
            mapping_keys_description='Activities',
            mapping={
            'logic': 'board game',
            'thinking': 'chess',
            'creative': 'painting'
        }),
        'learning more': LearningDecoder(),
        'no valid choice': None
    }

activities = ActivityDecoder().process(
    'Getting more education with something fun like numbers is what I like to do'
)

for activity in activities:
    print(activity)

Math Book

Note

All the examples are mostly using strings for values, but Decoder.mapping values can be any type of object so that you can get really creative when making decoders.