Skip to main content

Una libreria para construir redes bayesianas y realizar inferencia probabilística

Project description

Redes Bayesianas

Una librería para construir redes bayesianas y realizar inferencia probabilística.

Instalación

Con el manejador de paquetes pip:

  • pip install bayesian-networks-rey20074

Uso

from src.bayesian_networks_rey20074.BayesianNetwork import BayesianNetwork, Node

node_b = Node("b", 0.001)

node_e = Node("e", 0.002)

node_a = Node("a", multiple_parents=True)
node_a.add_connection_multiple_parents({"b": True, "e": True}, 0.95)
node_a.add_connection_multiple_parents({"b": True, "e": False}, 0.94)
node_a.add_connection_multiple_parents({"b": False, "e": True}, 0.29)
node_a.add_connection_multiple_parents({"b": False, "e": False}, 0.001)
node_a.add_connection("j", 0.9, True)
node_a.add_connection("j", 0.05, False)
node_a.add_connection("m", 0.7, True)
node_a.add_connection("m", 0.01, False)

node_j = Node("j")
node_m = Node("m")


network = BayesianNetwork()
network.add_node(node_a)
network.add_node(node_b)
network.add_node(node_e)
network.add_node(node_j)
network.add_node(node_m)
print('## GET PROBABILISTIC INFERENCE')
print(network.probabilistic_inference("m"))

print('\n## GET COMPACT REPRESENTATION')
representation = network.get_compact_representation()
print(representation)

print('## GET ELEMENTS USED FOR ALGORITHM')
collections = network.get_all_representations()
for x in collections:
    print(x)

print('\n## GET IS FULLY DESCRIBED')
desc = network.is_fully_described()

if (desc == True):
    print("Red Bayesiana Descriptiva")
else:
    print("Red Bayesiana No Descriptiva")

API

Se incluyen las siguientes clases

Clase Node

  • init(self, title: str, probability_of_success: float = None, multiple_parents: bool = False)
  • add_connection(self, next_node_title: str, probability_of_success: float, parent_was_succesful: bool)
  • get_children(self)
  • get_children_title(self)
  • delete_connection(self, node_title: str)
  • delete_connection(self, node_title: str, parent_was_succesful: bool)
  • add_connection_multiple_parents(self, parent_nodes: dict, probability_of_success: float)
  • edit_connection(self, node_title: str, probability_of_success: float)

Clase BayesianNetwork

  • init(self)
  • get_nodes(self)
  • get_node(self, node_title: str)
  • delete_node(self, node_title: str)
  • add_node(self, node: Node)
  • replace_node(self, node_title: str, new_node: Node)
  • get_parent(self, child_node_title: str)
  • get_parents(self, child_node_title: str)
  • one_parent_probabilistic_inference(self, node_title: str
  • multiply_list(myList: list)
  • multiple_parents_probabilistic_inference(self, node_title: str)
  • probabilistic_inference(self, node_title: str)
  • get_compact_representation(network)
  • is_fully_described(self)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

bayesian_networks_rey20074-0.1.5.tar.gz (5.0 kB view details)

Uploaded Source

Built Distribution

File details

Details for the file bayesian_networks_rey20074-0.1.5.tar.gz.

File metadata

File hashes

Hashes for bayesian_networks_rey20074-0.1.5.tar.gz
Algorithm Hash digest
SHA256 37800181cf4b259678c17e16c80e9f0df93293cb769acdee07e89275be0dc40b
MD5 a8b86ceb394c2dbac7b3448a38cf2b2e
BLAKE2b-256 756ba586c9dc018cad4d271b692f0f45e04f80a7b93ed897d7673b5dfc994ccc

See more details on using hashes here.

File details

Details for the file bayesian_networks_rey20074-0.1.5-py3-none-any.whl.

File metadata

File hashes

Hashes for bayesian_networks_rey20074-0.1.5-py3-none-any.whl
Algorithm Hash digest
SHA256 031064bc18f3b3471c43a9273b4bb8f411f9cc6a35410f4ae14fd5fcb6059fe5
MD5 fca096cd1bf84825677b1b6cc7812b5b
BLAKE2b-256 453e3a1ace9847ae60fd2da6f201d421ebb3a44f8d9bb3c9b108ba52f62b1a5f

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page