Python API
database
Persistant Profile

Documentation for AgentProfile, EnvironmentProfile, RelationshipProfile, and EnvironmentList Classes

Overview

This documentation provides details about four classes: AgentProfile, EnvironmentProfile, RelationshipProfile, and EnvironmentList. All classes are built using the pydantic and redis_om libraries to create data models that are stored and managed within a Redis database.

Dependencies

  • IntEnum from enum
  • Any from typing
  • root_validator from pydantic
  • JsonModel from redis_om
  • Field from redis_om.model.model

RelationshipType Enum

The RelationshipType enum defines various types of relationships between agents.

class RelationshipType(IntEnum):
    stranger = 0
    know_by_name = 1
    acquaintance = 2
    friend = 3
    romantic_relationship = 4
    family_member = 5

AgentProfile Class

The AgentProfile class is designed to store profile details of agents.

Class Definition

class AgentProfile(JsonModel):

Attributes

  • first_name (str): First name of the agent.
  • last_name (str): Last name of the agent.
  • age (int): Age of the agent. Default is 0.
  • occupation (str): Occupation of the agent. Default is an empty string.
  • gender (str): Gender of the agent. Default is an empty string.
  • gender_pronoun (str): Gender pronoun for the agent. Default is an empty string.
  • public_info (str): Public information about the agent. Default is an empty string.
  • big_five (str): Big Five personality traits. Default is an empty string.
  • moral_values (list[str]): List of moral values. Default is an empty list.
  • schwartz_personal_values (list[str]): Schwartz personal values. Default is an empty list.
  • personality_and_values (str): Combination of personality and values. Default is an empty string.
  • decision_making_style (str): Decision-making style of the agent. Default is an empty string.
  • secret (str): Secret information about the agent. Default is an empty string.
  • model_id (str): Associated model ID. Default is an empty string.
  • mbti (str): MBTI personality type. Default is an empty string.

EnvironmentProfile Class

The EnvironmentProfile class is designed to store profile details of environments.

Class Definition

class EnvironmentProfile(JsonModel):

Attributes

  • codename (str): Codename of the environment.
  • source (str): Source of the environment.
  • scenario (str): A concrete scenario where the social interaction takes place.
  • agent_goals (list[str]): Social goals of each agent.
  • relationship (RelationshipType): Relationship between the two agents.
  • age_constraint (Optional[str]): Age constraint ranges for agents.
  • occupation_constraint (Optional[str]): Occupation constraint for agents.
  • agent_constraint (Optional[list[list[str]]]): List of agent constraints.

RelationshipProfile Class

The RelationshipProfile class is designed to store relationship details between two agents.

Class Definition

class RelationshipProfile(JsonModel):

Attributes

  • agent_1_id (str): ID of the first agent.
  • agent_2_id (str): ID of the second agent.
  • relationship (RelationshipType): Relationship type between the two agents.
  • background_story (Optional[str]): Background story of the relationship.

EnvironmentList Class

The EnvironmentList class is designed to store lists of environments.

Class Definition

class EnvironmentList(JsonModel):

Attributes

  • name (str): Name of the environment list.
  • environments (list[str]): List of environments.
  • agent_index (Optional[list[str]]): List of agent indices.

Validators

the_length_agent_index_matches_environments

Ensures that the length of agent_index matches the length of environments.

Parameters

  • cls: The class being validated.
  • values (Any): Values provided to the class.

Returns

  • values (Any): Validated values.
@root_validator
def the_length_agent_index_matches_environments(cls, values: Any) -> Any:

Summary

  • The AgentProfile class manages profile details of individual agents.
  • The EnvironmentProfile class captures detailed information about environments.
  • The RelationshipProfile class stores relationship details between agents.
  • The EnvironmentList class manages lists of environments and validates the length of agent indices against environments.

All classes utilize the redis_om and pydantic libraries to facilitate storage, retrieval, and validation of data within a Redis database.

Example Usage

from database.persistant_profile import AgentProfile, EnvironmentProfile, RelationshipProfile, EnvironmentList
 
# Create an AgentProfile
agent_profile = AgentProfile(
    first_name="John",
    last_name="Doe",
    age=30,
    occupation="Engineer",
    gender="Male",
    gender_pronoun="He/Him",
    public_info="John is a software engineer.",
    big_five="Extroverted",
    moral_values=["Honesty", "Justice"],
    schwartz_personal_values=["Self-Direction", "Stimulation"],
    personality_and_values="John is a friendly and honest person.",
    decision_making_style="Analytical",
    secret="John is a secret agent.",
    model_id="1234567890",
    mbti="ENFP"
)
 
# Create an EnvironmentProfile
environment_profile = EnvironmentProfile(
    codename="Mission1",
    source="Scenario1",
    scenario="John and Jane are tasked with stopping a cyber attack.",
    agent_goals=["Stop the attack", "Save the data"],
    relationship=RelationshipType.acquaintance,
    age_constraint="25-35",
    occupation_constraint="Engineer",
)
 
# Create a RelationshipProfile
relationship_profile = RelationshipProfile(
    agent_1_id="1234567890",
    agent_2_id="0987654321",
    relationship=RelationshipType.friend,
    background_story="John and Jane have been friends since childhood."
)
 
# Create an EnvironmentList
environment_list = EnvironmentList(
    name="Mission1",
    environments=["550e8400-e29b-41d4-a716-446655440000", "f47ac10b-58cc-4372-a567-0e02b2c3d479"],
    agent_index=["1234567890", "0987654321"]
)

Save the profiles to Redis

agent_profile.save()
environment_profile.save()
relationship_profile.save()
environment_list.save()

Retrieve the profiles from Redis

retrieved_agent_profile = AgentProfile.find(AgentProfile.model_id == "1234567890").all()
retrieved_environment_profile = EnvironmentProfile.find(EnvironmentProfile.codename == "Mission1").all()
retrieved_relationship_profile = RelationshipProfile.find(RelationshipProfile.agent_1_id == "1234567890").all()
LogsEvaluators