models package

Submodules

models.base_langchain_model module

This module implements the base class for all LangChain models

class models.base_langchain_model.BaseLangChainModel(llm: LLM | SimpleChatModel = None, system_message: str | None = None, memory_kvargs: Dict[Any, Any] = None, **kvargs)

Bases: BaseLLMModel

This is the base class for all LangChain models

get_prompt_response(message: str, stream_handler: BaseCallbackHandler | None = None) → str

This method is used to get a response to a prompt

Parameters:

• message – The message to give to the LLM

• stream_handler – if passed response is streamed via handler

Returns:

The response from the LLM

get_prompt_response_without_memory(message: str, stream_handler: BaseCallbackHandler | None = None) → str

This method is used to get a response to a prompt The message is given to the LLM without any memory

Parameters:

• message – The message to give to the LLM

• stream_handler – if passed response is streamed via handler

Returns:

The response from the LLM

class models.base_langchain_model.StreamlitDisplayHandler(container: container, initial_text: str = '', display_method: str = 'markdown')

Bases: BaseCallbackHandler

This class is used to display the output of the LLM in streamlit

on_llm_end(response: str, **kwargs) → None

This method is used to display the output of the LLM in streamlit :param response: The response from the LLM :param **kwargs: The keyword arguments

on_llm_new_token(token: str, **kwargs) → None

This method is used to display the output of the LLM in streamlit

Parameters:

• token – The newly generated token

• **kwargs – The keyword arguments

models.base_model module

This file contains the base model class that all models should inherit from

class models.base_model.BaseLLMModel(system_message: str | None = None, **kvargs)

Bases: object

This is the base model class that all models should inherit from

add_ai_message(message: str)

This method is used to add an AI message to the conversation

Parameters:

message – The message to add to the conversation

add_user_message(message: str)

This method is used to add a user message to the conversation

Parameters:

message – The message to add to the conversation

get_messages() → List[Message]

This method is used to get the messages in the conversation

Returns:

The messages in the conversation

get_prompt_response(message: str, stream_handler: BaseCallbackHandler | None = None) → str

This method is used to get a response to a prompt

Parameters:

• message – The message to give to the LLM

• stream_handler – if passed response is streamed via handler

Returns:

The response from the LLM

get_prompt_response_without_memory(message: str, stream_handler: BaseCallbackHandler | None = None) → str

This method is used to get a response to a prompt The message is given to the LLM without any memory

Parameters:

• message – The message to give to the LLM

• stream_handler – if passed response is streamed via handler

Returns:

The response from the LLM

class models.base_model.FakeLLM(**kvargs)

Bases: object

This is a fake LLM class that is used for testing

give_response_to_prompt(messages: List[Message], system_prompt: str) → str

This method is used to give a response to a prompt

Parameters:

• messages – The messages to give to the LLM

• system_prompt – The system prompt to give to the LLM

Returns:

The response from the LLM

models.chat_gpt module

ChatGPT is a class that inherits from BaseLangChainModel and uses the ChatOpenAI class to generate text.

class models.chat_gpt.ChatGPT(system_message: str | None = None, memory_kvargs: Dict[Any, Any] = None, **kvargs)

Bases: BaseLangChainModel

This is the class for the ChatGPT model

models.llama2 module

LLAMA2 model

class models.llama2.LLAMA2(system_message: str | None = None, memory_kvargs: Dict[Any, Any] = None, **kvargs)

Bases: BaseLangChainModel

This is the class for the LLAMA2 model

models.message module

This is the module for the message data class

class models.message.Message(*, message: str, message_type: Literal['SYSTEM', 'USER', 'AI'], timestamp: datetime = None)

Bases: BaseModel

This is the class for a message

message: str

message_type: Literal['SYSTEM', 'USER', 'AI']

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'message': FieldInfo(annotation=str, required=True, description='The message'), 'message_type': FieldInfo(annotation=Literal[str, str, str], required=True, description='The type of message'), 'timestamp': FieldInfo(annotation=datetime, required=False, default_factory=builtin_function_or_method, description='The timestamp of the message')}

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

timestamp: datetime

models.meta_info module

This module contains the meta information for a model

class models.meta_info.ModelMetaInfo(*, base_dir: str, key: str | None = None, name: str, description: str, llm_model_file: str | None = None, llm_model_class: str | None = None, supports_stream: bool = None, icon: str = 'llm_model.png', system_message: str | None = 'You are a helpful assistant that responds in Markdown format.', llm_arguments: Dict[str, Any] = {}, memory_arguments: Dict[str, Any] = {}, required_llm_arguments: Dict[str, Literal['STRING', 'INT', 'FLOAT', 'BOOL', 'SECRET_STRING', 'LONG_STRING']] = {}, inherits_from: str | None = None, is_persistent: bool = True)

Bases: StreamLitPydanticModel

This class is used to store the meta information for a model

add_conversation(conversation: Conversation) → None

This method is used to add a conversation

Parameters:

conversation – The conversation

additional_custom_fields_to_edit() → list[str | FormatOption]

This method is used to get the additional fields to show

Returns:

The additional fields to show

additional_custom_fields_to_show() → list[str | FormatOption]

This method is used to get the additional fields to show

Returns:

The additional fields to show

base_dir: str

description: str

classmethod fields_to_edit() → list[str | Dict[str, Dict[str, FormatOption]]]

This method is used to get the fields to edit

Returns:

The fields to edit

classmethod fields_to_show() → list[str | FormatOption]

This method is used to get the fields to show

Returns:

The fields to show

get_additional_custom_field_value(field_name: str) → Any

This method is used to get the value for custom fields

Parameters:

field_name – The field name

Returns:

The field value

get_conversations() → list[Conversation]

This method is used to get the conversations

Returns:

The conversations

icon: str

inherits_from: str | None

is_persistent: bool

key: str | None

llm_arguments: Dict[str, Any]

llm_model_class: str | None

llm_model_file: str | None

memory_arguments: Dict[str, Any]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_fields: ClassVar[dict[str, FieldInfo]] = {'base_dir': FieldInfo(annotation=str, required=True, alias_priority=2, validation_alias=AliasChoices(choices=['base_dir', 'BaseDir']), description='The base dir of the app'), 'description': FieldInfo(annotation=str, required=True, alias_priority=2, validation_alias=AliasChoices(choices=['description', 'Description']), description='The description of the model', metadata=[MinLen(min_length=20)]), 'icon': FieldInfo(annotation=str, required=False, default='llm_model.png', alias_priority=2, validation_alias=AliasChoices(choices=['icon', 'Icon']), description='The icon for the model'), 'inherits_from': FieldInfo(annotation=Union[str, NoneType], required=False, alias_priority=2, validation_alias=AliasChoices(choices=['inherits_from', 'InheritsFrom']), description='The model to inherit from'), 'is_persistent': FieldInfo(annotation=bool, required=False, default=True, alias_priority=2, validation_alias=AliasChoices(choices=['is_persistent', 'IsPersistent']), description='Whether the model is persistent'), 'key': FieldInfo(annotation=Union[str, NoneType], required=False, description='The key for the model'), 'llm_arguments': FieldInfo(annotation=Dict[str, Any], required=False, default={}, alias_priority=2, validation_alias=AliasChoices(choices=['llm_arguments', 'LLMArguments']), description='The arguments for the LLM Model'), 'llm_model_class': FieldInfo(annotation=Union[str, NoneType], required=False, alias_priority=2, validation_alias=AliasChoices(choices=['llm_model_class', 'LLMModelClass']), description='The model class'), 'llm_model_file': FieldInfo(annotation=Union[str, NoneType], required=False, alias_priority=2, validation_alias=AliasChoices(choices=['llm_model_file', 'LLMModelFile']), description='The model file'), 'memory_arguments': FieldInfo(annotation=Dict[str, Any], required=False, default={}, alias_priority=2, validation_alias=AliasChoices(choices=['memory_arguments', 'MemoryArguments']), description='The arguments for the chat memory module'), 'name': FieldInfo(annotation=str, required=True, alias_priority=2, validation_alias=AliasChoices(choices=['name', 'Name']), description='The name of the model', metadata=[MinLen(min_length=3)]), 'required_llm_arguments': FieldInfo(annotation=Dict[str, Literal[str, str, str, str, str, str]], required=False, default={}, alias_priority=2, validation_alias=AliasChoices(choices=['required_llm_arguments', 'RequiredLLMArguments']), description='The required arguments for the model'), 'supports_stream': FieldInfo(annotation=bool, required=False, alias_priority=2, validation_alias=AliasChoices(choices=['supports_stream', 'SupportsStream']), description='Whether the model supports streaming'), 'system_message': FieldInfo(annotation=Union[str, NoneType], required=False, default='You are a helpful assistant that responds in Markdown format.', alias_priority=2, validation_alias=AliasChoices(choices=['system_message', 'SystemMessage']), description='The system message for the model')}

Metadata about the fields defined on the model, mapping of field names to [FieldInfo][pydantic.fields.FieldInfo].

This replaces Model.__fields__ from Pydantic V1.

model_post_init(__context: Any) → None

This function is meant to behave like a BaseModel method to initialise private attributes.

It takes context as an argument since that’s what pydantic-core passes when calling it.

Parameters:

• self – The BaseModel instance.

• __context – The context.

name: str

required_llm_arguments: Dict[str, Literal['STRING', 'INT', 'FLOAT', 'BOOL', 'SECRET_STRING', 'LONG_STRING']]

set_additional_custom_field_value(field_name: str, field_value: Any)

This method is used to set the value for custom fields

Parameters:

• field_name – The field name

• field_value – The field value

set_value_for_custom_fields(field_name: str, field_value: Any)

This method is used to set the value for custom fields

Parameters:

• field_name – The field name

• field_value – The field value

supports_stream: bool

system_message: str | None

classmethod validate_base_dir(value: str) → str

Validates the base dir

Parameters:

value – The value of the base dir

Returns:

The value of the base dir

Raises:

ValueError – If the base dir is not found

classmethod validate_icon(value: str, field_info: FieldValidationInfo) → str

Validates the icon

Parameters:

• value – The value of the icon

• field_info – The field info object from pydantic.

Returns:

The value of the icon

Return type:

str

Raises:

ValueError – If the icon is not found

classmethod validate_model(data: Dict, field_info: FieldValidationInfo) → Dict

Validates the model

Parameters:

• data – The data of the model

• field_info – The field info object from pydantic.

Returns:

The processed and validated data for the model

Raises:

ValueError – If the model is invalid

Module contents