Thinkflux Client

TFClient is the main class utilized to interface with a Thinkflux reasoning agent. It provides several helper functions and API calls to aid in data input, agent configuration, and debugging.

Quickstart

class ia.gaius.thinkflux_client.TFClient(tf_info: dict, verify: bool = True)

Bases: object

Thinkflux Client Wrapper (interface to TF REST API)

__init__(tf_info: dict, verify: bool = True) → None
add_interface_nodes(agent_config: dict) → dict

Add an interface node from GAIuS agent to Thinkflux

Parameters:

  • agent_config (dict) – dictionary with configuration details for interface node

  • fields (agent_config dict needs the following) –

  • { – “name”: str, “domain: str, “secure”: bool, “api_key”: str, “interface_nodes”: List[str]

  • } –

add_schema_symbol_information() → Response
bootstrap_concepts(hierarchy: bool = False, labelled: bool = False, use_labels_for_concepts: bool = False, use_labels_for_hierarchy: bool = False, subsequence_clustering: bool = False, **kwargs) → Response

_summary_

Parameters:

  • hierarchy (bool, optional) – Whether to generate a hierarchy of concepts. Defaults to False.

  • labelled (bool, optional) – If the dataset contains labels. Defaults to False.

  • use_labels_for_concepts (bool, optional) – Whether labels should be used for concept names. Defaults to False.

  • use_labels_for_hierarchy (bool, optional) – Make a hierarchy using labels. Defaults to False.

  • subsequence_clustering (bool, optional) – Perform subsequence clustering when generating concepts. Defaults to False.

  • **kwargs – Other arguments to pass

Returns:

Response object from requests

Return type:

Response

clear_all_emotives() → dict
clear_all_kbs() → dict
clear_concepts_and_instances() → dict

Remove all Concepts and instances from Thinkflux

Returns:

_description_

Return type:

dict

clear_wm() → dict
delete_interface_nodes(data: dict) → Response
dreamer(data: dict) → Response
evaluate_world(data: dict) → dict
get_concept(concept_name: str) → dict

Retrieve a single concept from Thinkflux by concept name

Parameters:

concept_name (str) – Name of the ooncept to retrieve

get_concepts() → dict

Retrieve all concepts from Thinkflux

get_instances() → dict

Retrieve current instances from Thinkflux

get_model_association_networks() → Response
get_plan() → dict
get_rules_kb() → dict
get_symbolic_goal() → dict
investigate(data: dict) → Response
list_interface_nodes() → dict
load_schema_base(schema_base: dict) → Response

Load a SchemaBase on Thinkflux, from a dictionary object

Parameters:

schema_base (dict) – the SchemaBase to load

Returns:

requests Response object

Return type:

Response

observe(data: dict) → dict

Observe data on Thinkflux

Parameters:

data (dict) – data percept to observe

ping() → str

Ensure connection can be established with Thinkflux. Sanity check.

Raises:

TFQueryError – _description_

Returns:

_description_

Return type:

str

set_rules_kb(rules_kb: dict) → dict
set_symbolic_goal(symbolic_goal: dict) → dict
set_verify(verify: bool)
show_status() → dict
timer() → Response
update_schema(data: dict) → Response