refactor(api/core): Improve type hints and apply ruff formatter in agent runner and model manager. (#8166)

This commit is contained in:
-LAN- 2024-09-10 15:00:25 +08:00 committed by GitHub
parent af92f19291
commit ed37439ef7
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 199 additions and 197 deletions

View File

@ -1,6 +1,7 @@
import json
import logging
import uuid
from collections.abc import Mapping, Sequence
from datetime import datetime, timezone
from typing import Optional, Union, cast
@ -45,22 +46,25 @@ from models.tools import ToolConversationVariables
logger = logging.getLogger(__name__)
class BaseAgentRunner(AppRunner):
def __init__(self, tenant_id: str,
application_generate_entity: AgentChatAppGenerateEntity,
conversation: Conversation,
app_config: AgentChatAppConfig,
model_config: ModelConfigWithCredentialsEntity,
config: AgentEntity,
queue_manager: AppQueueManager,
message: Message,
user_id: str,
memory: Optional[TokenBufferMemory] = None,
prompt_messages: Optional[list[PromptMessage]] = None,
variables_pool: Optional[ToolRuntimeVariablePool] = None,
db_variables: Optional[ToolConversationVariables] = None,
model_instance: ModelInstance = None
) -> None:
def __init__(
self,
tenant_id: str,
application_generate_entity: AgentChatAppGenerateEntity,
conversation: Conversation,
app_config: AgentChatAppConfig,
model_config: ModelConfigWithCredentialsEntity,
config: AgentEntity,
queue_manager: AppQueueManager,
message: Message,
user_id: str,
memory: Optional[TokenBufferMemory] = None,
prompt_messages: Optional[list[PromptMessage]] = None,
variables_pool: Optional[ToolRuntimeVariablePool] = None,
db_variables: Optional[ToolConversationVariables] = None,
model_instance: ModelInstance = None,
) -> None:
"""
Agent runner
:param tenant_id: tenant id
@ -88,9 +92,7 @@ class BaseAgentRunner(AppRunner):
self.message = message
self.user_id = user_id
self.memory = memory
self.history_prompt_messages = self.organize_agent_history(
prompt_messages=prompt_messages or []
)
self.history_prompt_messages = self.organize_agent_history(prompt_messages=prompt_messages or [])
self.variables_pool = variables_pool
self.db_variables_pool = db_variables
self.model_instance = model_instance
@ -111,12 +113,16 @@ class BaseAgentRunner(AppRunner):
retrieve_config=app_config.dataset.retrieve_config if app_config.dataset else None,
return_resource=app_config.additional_features.show_retrieve_source,
invoke_from=application_generate_entity.invoke_from,
hit_callback=hit_callback
hit_callback=hit_callback,
)
# get how many agent thoughts have been created
self.agent_thought_count = db.session.query(MessageAgentThought).filter(
MessageAgentThought.message_id == self.message.id,
).count()
self.agent_thought_count = (
db.session.query(MessageAgentThought)
.filter(
MessageAgentThought.message_id == self.message.id,
)
.count()
)
db.session.close()
# check if model supports stream tool call
@ -135,25 +141,26 @@ class BaseAgentRunner(AppRunner):
self.query = None
self._current_thoughts: list[PromptMessage] = []
def _repack_app_generate_entity(self, app_generate_entity: AgentChatAppGenerateEntity) \
-> AgentChatAppGenerateEntity:
def _repack_app_generate_entity(
self, app_generate_entity: AgentChatAppGenerateEntity
) -> AgentChatAppGenerateEntity:
"""
Repack app generate entity
"""
if app_generate_entity.app_config.prompt_template.simple_prompt_template is None:
app_generate_entity.app_config.prompt_template.simple_prompt_template = ''
app_generate_entity.app_config.prompt_template.simple_prompt_template = ""
return app_generate_entity
def _convert_tool_to_prompt_message_tool(self, tool: AgentToolEntity) -> tuple[PromptMessageTool, Tool]:
"""
convert tool to prompt message tool
convert tool to prompt message tool
"""
tool_entity = ToolManager.get_agent_tool_runtime(
tenant_id=self.tenant_id,
app_id=self.app_config.app_id,
agent_tool=tool,
invoke_from=self.application_generate_entity.invoke_from
invoke_from=self.application_generate_entity.invoke_from,
)
tool_entity.load_variables(self.variables_pool)
@ -164,7 +171,7 @@ class BaseAgentRunner(AppRunner):
"type": "object",
"properties": {},
"required": [],
}
},
)
parameters = tool_entity.get_all_runtime_parameters()
@ -177,19 +184,19 @@ class BaseAgentRunner(AppRunner):
if parameter.type == ToolParameter.ToolParameterType.SELECT:
enum = [option.value for option in parameter.options]
message_tool.parameters['properties'][parameter.name] = {
message_tool.parameters["properties"][parameter.name] = {
"type": parameter_type,
"description": parameter.llm_description or '',
"description": parameter.llm_description or "",
}
if len(enum) > 0:
message_tool.parameters['properties'][parameter.name]['enum'] = enum
message_tool.parameters["properties"][parameter.name]["enum"] = enum
if parameter.required:
message_tool.parameters['required'].append(parameter.name)
message_tool.parameters["required"].append(parameter.name)
return message_tool, tool_entity
def _convert_dataset_retriever_tool_to_prompt_message_tool(self, tool: DatasetRetrieverTool) -> PromptMessageTool:
"""
convert dataset retriever tool to prompt message tool
@ -201,24 +208,24 @@ class BaseAgentRunner(AppRunner):
"type": "object",
"properties": {},
"required": [],
}
},
)
for parameter in tool.get_runtime_parameters():
parameter_type = 'string'
prompt_tool.parameters['properties'][parameter.name] = {
parameter_type = "string"
prompt_tool.parameters["properties"][parameter.name] = {
"type": parameter_type,
"description": parameter.llm_description or '',
"description": parameter.llm_description or "",
}
if parameter.required:
if parameter.name not in prompt_tool.parameters['required']:
prompt_tool.parameters['required'].append(parameter.name)
if parameter.name not in prompt_tool.parameters["required"]:
prompt_tool.parameters["required"].append(parameter.name)
return prompt_tool
def _init_prompt_tools(self) -> tuple[dict[str, Tool], list[PromptMessageTool]]:
def _init_prompt_tools(self) -> tuple[Mapping[str, Tool], Sequence[PromptMessageTool]]:
"""
Init tools
"""
@ -261,51 +268,51 @@ class BaseAgentRunner(AppRunner):
enum = []
if parameter.type == ToolParameter.ToolParameterType.SELECT:
enum = [option.value for option in parameter.options]
prompt_tool.parameters['properties'][parameter.name] = {
prompt_tool.parameters["properties"][parameter.name] = {
"type": parameter_type,
"description": parameter.llm_description or '',
"description": parameter.llm_description or "",
}
if len(enum) > 0:
prompt_tool.parameters['properties'][parameter.name]['enum'] = enum
prompt_tool.parameters["properties"][parameter.name]["enum"] = enum
if parameter.required:
if parameter.name not in prompt_tool.parameters['required']:
prompt_tool.parameters['required'].append(parameter.name)
if parameter.name not in prompt_tool.parameters["required"]:
prompt_tool.parameters["required"].append(parameter.name)
return prompt_tool
def create_agent_thought(self, message_id: str, message: str,
tool_name: str, tool_input: str, messages_ids: list[str]
) -> MessageAgentThought:
def create_agent_thought(
self, message_id: str, message: str, tool_name: str, tool_input: str, messages_ids: list[str]
) -> MessageAgentThought:
"""
Create agent thought
"""
thought = MessageAgentThought(
message_id=message_id,
message_chain_id=None,
thought='',
thought="",
tool=tool_name,
tool_labels_str='{}',
tool_meta_str='{}',
tool_labels_str="{}",
tool_meta_str="{}",
tool_input=tool_input,
message=message,
message_token=0,
message_unit_price=0,
message_price_unit=0,
message_files=json.dumps(messages_ids) if messages_ids else '',
answer='',
observation='',
message_files=json.dumps(messages_ids) if messages_ids else "",
answer="",
observation="",
answer_token=0,
answer_unit_price=0,
answer_price_unit=0,
tokens=0,
total_price=0,
position=self.agent_thought_count + 1,
currency='USD',
currency="USD",
latency=0,
created_by_role='account',
created_by_role="account",
created_by=self.user_id,
)
@ -318,22 +325,22 @@ class BaseAgentRunner(AppRunner):
return thought
def save_agent_thought(self,
agent_thought: MessageAgentThought,
tool_name: str,
tool_input: Union[str, dict],
thought: str,
observation: Union[str, dict],
tool_invoke_meta: Union[str, dict],
answer: str,
messages_ids: list[str],
llm_usage: LLMUsage = None) -> MessageAgentThought:
def save_agent_thought(
self,
agent_thought: MessageAgentThought,
tool_name: str,
tool_input: Union[str, dict],
thought: str,
observation: Union[str, dict],
tool_invoke_meta: Union[str, dict],
answer: str,
messages_ids: list[str],
llm_usage: LLMUsage = None,
) -> MessageAgentThought:
"""
Save agent thought
"""
agent_thought = db.session.query(MessageAgentThought).filter(
MessageAgentThought.id == agent_thought.id
).first()
agent_thought = db.session.query(MessageAgentThought).filter(MessageAgentThought.id == agent_thought.id).first()
if thought is not None:
agent_thought.thought = thought
@ -356,7 +363,7 @@ class BaseAgentRunner(AppRunner):
observation = json.dumps(observation, ensure_ascii=False)
except Exception as e:
observation = json.dumps(observation)
agent_thought.observation = observation
if answer is not None:
@ -364,7 +371,7 @@ class BaseAgentRunner(AppRunner):
if messages_ids is not None and len(messages_ids) > 0:
agent_thought.message_files = json.dumps(messages_ids)
if llm_usage:
agent_thought.message_token = llm_usage.prompt_tokens
agent_thought.message_price_unit = llm_usage.prompt_price_unit
@ -377,7 +384,7 @@ class BaseAgentRunner(AppRunner):
# check if tool labels is not empty
labels = agent_thought.tool_labels or {}
tools = agent_thought.tool.split(';') if agent_thought.tool else []
tools = agent_thought.tool.split(";") if agent_thought.tool else []
for tool in tools:
if not tool:
continue
@ -386,7 +393,7 @@ class BaseAgentRunner(AppRunner):
if tool_label:
labels[tool] = tool_label.to_dict()
else:
labels[tool] = {'en_US': tool, 'zh_Hans': tool}
labels[tool] = {"en_US": tool, "zh_Hans": tool}
agent_thought.tool_labels_str = json.dumps(labels)
@ -401,14 +408,18 @@ class BaseAgentRunner(AppRunner):
db.session.commit()
db.session.close()
def update_db_variables(self, tool_variables: ToolRuntimeVariablePool, db_variables: ToolConversationVariables):
"""
convert tool variables to db variables
"""
db_variables = db.session.query(ToolConversationVariables).filter(
ToolConversationVariables.conversation_id == self.message.conversation_id,
).first()
db_variables = (
db.session.query(ToolConversationVariables)
.filter(
ToolConversationVariables.conversation_id == self.message.conversation_id,
)
.first()
)
db_variables.updated_at = datetime.now(timezone.utc).replace(tzinfo=None)
db_variables.variables_str = json.dumps(jsonable_encoder(tool_variables.pool))
@ -425,9 +436,14 @@ class BaseAgentRunner(AppRunner):
if isinstance(prompt_message, SystemPromptMessage):
result.append(prompt_message)
messages: list[Message] = db.session.query(Message).filter(
Message.conversation_id == self.message.conversation_id,
).order_by(Message.created_at.asc()).all()
messages: list[Message] = (
db.session.query(Message)
.filter(
Message.conversation_id == self.message.conversation_id,
)
.order_by(Message.created_at.asc())
.all()
)
for message in messages:
if message.id == self.message.id:
@ -439,13 +455,13 @@ class BaseAgentRunner(AppRunner):
for agent_thought in agent_thoughts:
tools = agent_thought.tool
if tools:
tools = tools.split(';')
tools = tools.split(";")
tool_calls: list[AssistantPromptMessage.ToolCall] = []
tool_call_response: list[ToolPromptMessage] = []
try:
tool_inputs = json.loads(agent_thought.tool_input)
except Exception as e:
tool_inputs = { tool: {} for tool in tools }
tool_inputs = {tool: {} for tool in tools}
try:
tool_responses = json.loads(agent_thought.observation)
except Exception as e:
@ -454,27 +470,33 @@ class BaseAgentRunner(AppRunner):
for tool in tools:
# generate a uuid for tool call
tool_call_id = str(uuid.uuid4())
tool_calls.append(AssistantPromptMessage.ToolCall(
id=tool_call_id,
type='function',
function=AssistantPromptMessage.ToolCall.ToolCallFunction(
name=tool,
arguments=json.dumps(tool_inputs.get(tool, {})),
tool_calls.append(
AssistantPromptMessage.ToolCall(
id=tool_call_id,
type="function",
function=AssistantPromptMessage.ToolCall.ToolCallFunction(
name=tool,
arguments=json.dumps(tool_inputs.get(tool, {})),
),
)
))
tool_call_response.append(ToolPromptMessage(
content=tool_responses.get(tool, agent_thought.observation),
name=tool,
tool_call_id=tool_call_id,
))
)
tool_call_response.append(
ToolPromptMessage(
content=tool_responses.get(tool, agent_thought.observation),
name=tool,
tool_call_id=tool_call_id,
)
)
result.extend([
AssistantPromptMessage(
content=agent_thought.thought,
tool_calls=tool_calls,
),
*tool_call_response
])
result.extend(
[
AssistantPromptMessage(
content=agent_thought.thought,
tool_calls=tool_calls,
),
*tool_call_response,
]
)
if not tools:
result.append(AssistantPromptMessage(content=agent_thought.thought))
else:
@ -496,10 +518,7 @@ class BaseAgentRunner(AppRunner):
file_extra_config = FileUploadConfigManager.convert(message.app_model_config.to_dict())
if file_extra_config:
file_objs = message_file_parser.transform_message_files(
files,
file_extra_config
)
file_objs = message_file_parser.transform_message_files(files, file_extra_config)
else:
file_objs = []

View File

@ -1,6 +1,6 @@
import logging
import os
from collections.abc import Callable, Generator
from collections.abc import Callable, Generator, Sequence
from typing import IO, Optional, Union, cast
from core.entities.provider_configuration import ProviderConfiguration, ProviderModelBundle
@ -41,7 +41,7 @@ class ModelInstance:
configuration=provider_model_bundle.configuration,
model_type=provider_model_bundle.model_type_instance.model_type,
model=model,
credentials=self.credentials
credentials=self.credentials,
)
@staticmethod
@ -54,10 +54,7 @@ class ModelInstance:
"""
configuration = provider_model_bundle.configuration
model_type = provider_model_bundle.model_type_instance.model_type
credentials = configuration.get_current_credentials(
model_type=model_type,
model=model
)
credentials = configuration.get_current_credentials(model_type=model_type, model=model)
if credentials is None:
raise ProviderTokenNotInitError(f"Model {model} credentials is not initialized.")
@ -65,10 +62,9 @@ class ModelInstance:
return credentials
@staticmethod
def _get_load_balancing_manager(configuration: ProviderConfiguration,
model_type: ModelType,
model: str,
credentials: dict) -> Optional["LBModelManager"]:
def _get_load_balancing_manager(
configuration: ProviderConfiguration, model_type: ModelType, model: str, credentials: dict
) -> Optional["LBModelManager"]:
"""
Get load balancing model credentials
:param configuration: provider configuration
@ -81,8 +77,7 @@ class ModelInstance:
current_model_setting = None
# check if model is disabled by admin
for model_setting in configuration.model_settings:
if (model_setting.model_type == model_type
and model_setting.model == model):
if model_setting.model_type == model_type and model_setting.model == model:
current_model_setting = model_setting
break
@ -95,17 +90,23 @@ class ModelInstance:
model_type=model_type,
model=model,
load_balancing_configs=current_model_setting.load_balancing_configs,
managed_credentials=credentials if configuration.custom_configuration.provider else None
managed_credentials=credentials if configuration.custom_configuration.provider else None,
)
return lb_model_manager
return None
def invoke_llm(self, prompt_messages: list[PromptMessage], model_parameters: Optional[dict] = None,
tools: Optional[list[PromptMessageTool]] = None, stop: Optional[list[str]] = None,
stream: bool = True, user: Optional[str] = None, callbacks: Optional[list[Callback]] = None) \
-> Union[LLMResult, Generator]:
def invoke_llm(
self,
prompt_messages: list[PromptMessage],
model_parameters: Optional[dict] = None,
tools: Sequence[PromptMessageTool] | None = None,
stop: Optional[list[str]] = None,
stream: bool = True,
user: Optional[str] = None,
callbacks: Optional[list[Callback]] = None,
) -> Union[LLMResult, Generator]:
"""
Invoke large language model
@ -132,11 +133,12 @@ class ModelInstance:
stop=stop,
stream=stream,
user=user,
callbacks=callbacks
callbacks=callbacks,
)
def get_llm_num_tokens(self, prompt_messages: list[PromptMessage],
tools: Optional[list[PromptMessageTool]] = None) -> int:
def get_llm_num_tokens(
self, prompt_messages: list[PromptMessage], tools: Optional[list[PromptMessageTool]] = None
) -> int:
"""
Get number of tokens for llm
@ -153,11 +155,10 @@ class ModelInstance:
model=self.model,
credentials=self.credentials,
prompt_messages=prompt_messages,
tools=tools
tools=tools,
)
def invoke_text_embedding(self, texts: list[str], user: Optional[str] = None) \
-> TextEmbeddingResult:
def invoke_text_embedding(self, texts: list[str], user: Optional[str] = None) -> TextEmbeddingResult:
"""
Invoke large language model
@ -174,7 +175,7 @@ class ModelInstance:
model=self.model,
credentials=self.credentials,
texts=texts,
user=user
user=user,
)
def get_text_embedding_num_tokens(self, texts: list[str]) -> int:
@ -192,13 +193,17 @@ class ModelInstance:
function=self.model_type_instance.get_num_tokens,
model=self.model,
credentials=self.credentials,
texts=texts
texts=texts,
)
def invoke_rerank(self, query: str, docs: list[str], score_threshold: Optional[float] = None,
top_n: Optional[int] = None,
user: Optional[str] = None) \
-> RerankResult:
def invoke_rerank(
self,
query: str,
docs: list[str],
score_threshold: Optional[float] = None,
top_n: Optional[int] = None,
user: Optional[str] = None,
) -> RerankResult:
"""
Invoke rerank model
@ -221,11 +226,10 @@ class ModelInstance:
docs=docs,
score_threshold=score_threshold,
top_n=top_n,
user=user
user=user,
)
def invoke_moderation(self, text: str, user: Optional[str] = None) \
-> bool:
def invoke_moderation(self, text: str, user: Optional[str] = None) -> bool:
"""
Invoke moderation model
@ -242,11 +246,10 @@ class ModelInstance:
model=self.model,
credentials=self.credentials,
text=text,
user=user
user=user,
)
def invoke_speech2text(self, file: IO[bytes], user: Optional[str] = None) \
-> str:
def invoke_speech2text(self, file: IO[bytes], user: Optional[str] = None) -> str:
"""
Invoke large language model
@ -263,11 +266,10 @@ class ModelInstance:
model=self.model,
credentials=self.credentials,
file=file,
user=user
user=user,
)
def invoke_tts(self, content_text: str, tenant_id: str, voice: str, user: Optional[str] = None) \
-> str:
def invoke_tts(self, content_text: str, tenant_id: str, voice: str, user: Optional[str] = None) -> str:
"""
Invoke large language tts model
@ -288,7 +290,7 @@ class ModelInstance:
content_text=content_text,
user=user,
tenant_id=tenant_id,
voice=voice
voice=voice,
)
def _round_robin_invoke(self, function: Callable, *args, **kwargs):
@ -312,8 +314,8 @@ class ModelInstance:
raise last_exception
try:
if 'credentials' in kwargs:
del kwargs['credentials']
if "credentials" in kwargs:
del kwargs["credentials"]
return function(*args, **kwargs, credentials=lb_config.credentials)
except InvokeRateLimitError as e:
# expire in 60 seconds
@ -340,9 +342,7 @@ class ModelInstance:
self.model_type_instance = cast(TTSModel, self.model_type_instance)
return self.model_type_instance.get_tts_model_voices(
model=self.model,
credentials=self.credentials,
language=language
model=self.model, credentials=self.credentials, language=language
)
@ -363,9 +363,7 @@ class ModelManager:
return self.get_default_model_instance(tenant_id, model_type)
provider_model_bundle = self._provider_manager.get_provider_model_bundle(
tenant_id=tenant_id,
provider=provider,
model_type=model_type
tenant_id=tenant_id, provider=provider, model_type=model_type
)
return ModelInstance(provider_model_bundle, model)
@ -386,10 +384,7 @@ class ModelManager:
:param model_type: model type
:return:
"""
default_model_entity = self._provider_manager.get_default_model(
tenant_id=tenant_id,
model_type=model_type
)
default_model_entity = self._provider_manager.get_default_model(tenant_id=tenant_id, model_type=model_type)
if not default_model_entity:
raise ProviderTokenNotInitError(f"Default model not found for {model_type}")
@ -398,17 +393,20 @@ class ModelManager:
tenant_id=tenant_id,
provider=default_model_entity.provider.provider,
model_type=model_type,
model=default_model_entity.model
model=default_model_entity.model,
)
class LBModelManager:
def __init__(self, tenant_id: str,
provider: str,
model_type: ModelType,
model: str,
load_balancing_configs: list[ModelLoadBalancingConfiguration],
managed_credentials: Optional[dict] = None) -> None:
def __init__(
self,
tenant_id: str,
provider: str,
model_type: ModelType,
model: str,
load_balancing_configs: list[ModelLoadBalancingConfiguration],
managed_credentials: Optional[dict] = None,
) -> None:
"""
Load balancing model manager
:param tenant_id: tenant_id
@ -439,10 +437,7 @@ class LBModelManager:
:return:
"""
cache_key = "model_lb_index:{}:{}:{}:{}".format(
self._tenant_id,
self._provider,
self._model_type.value,
self._model
self._tenant_id, self._provider, self._model_type.value, self._model
)
cooldown_load_balancing_configs = []
@ -473,10 +468,12 @@ class LBModelManager:
continue
if bool(os.environ.get("DEBUG", 'False').lower() == 'true'):
logger.info(f"Model LB\nid: {config.id}\nname:{config.name}\n"
f"tenant_id: {self._tenant_id}\nprovider: {self._provider}\n"
f"model_type: {self._model_type.value}\nmodel: {self._model}")
if bool(os.environ.get("DEBUG", "False").lower() == "true"):
logger.info(
f"Model LB\nid: {config.id}\nname:{config.name}\n"
f"tenant_id: {self._tenant_id}\nprovider: {self._provider}\n"
f"model_type: {self._model_type.value}\nmodel: {self._model}"
)
return config
@ -490,14 +487,10 @@ class LBModelManager:
:return:
"""
cooldown_cache_key = "model_lb_index:cooldown:{}:{}:{}:{}:{}".format(
self._tenant_id,
self._provider,
self._model_type.value,
self._model,
config.id
self._tenant_id, self._provider, self._model_type.value, self._model, config.id
)
redis_client.setex(cooldown_cache_key, expire, 'true')
redis_client.setex(cooldown_cache_key, expire, "true")
def in_cooldown(self, config: ModelLoadBalancingConfiguration) -> bool:
"""
@ -506,11 +499,7 @@ class LBModelManager:
:return:
"""
cooldown_cache_key = "model_lb_index:cooldown:{}:{}:{}:{}:{}".format(
self._tenant_id,
self._provider,
self._model_type.value,
self._model,
config.id
self._tenant_id, self._provider, self._model_type.value, self._model, config.id
)
res = redis_client.exists(cooldown_cache_key)
@ -518,11 +507,9 @@ class LBModelManager:
return res
@staticmethod
def get_config_in_cooldown_and_ttl(tenant_id: str,
provider: str,
model_type: ModelType,
model: str,
config_id: str) -> tuple[bool, int]:
def get_config_in_cooldown_and_ttl(
tenant_id: str, provider: str, model_type: ModelType, model: str, config_id: str
) -> tuple[bool, int]:
"""
Get model load balancing config is in cooldown and ttl
:param tenant_id: workspace id
@ -533,11 +520,7 @@ class LBModelManager:
:return:
"""
cooldown_cache_key = "model_lb_index:cooldown:{}:{}:{}:{}:{}".format(
tenant_id,
provider,
model_type.value,
model,
config_id
tenant_id, provider, model_type.value, model, config_id
)
ttl = redis_client.ttl(cooldown_cache_key)