Weyaxi's picture
Duplicate from Weyaxi/commit-trash-huggingface-spaces-codes
42472b3
raw
history blame
9.6 kB
import re
from typing import List, Union
from langchain.chains import LLMChain
from langchain.agents import Tool, LLMSingleActionAgent, AgentExecutor, AgentOutputParser
from langchain.schema import AgentAction, AgentFinish
from langchain.agents import initialize_agent
from langchain.prompts import StringPromptTemplate
from agents.promopts import code_generate_agent_template
from agents.tools.smart_domain.api_layer_code_tool import apiLayerCodeGenerator
from agents.tools.smart_domain.domain_layer_code_tool import domainLayerCodeGenerator
from agents.tools.smart_domain.entity import entityCodeGenerator
from agents.tools.smart_domain.association import associationCodeGenerator
from agents.tools.smart_domain.db_entity_repository import dbEntityRepositoryCodeGenerator
from agents.tools.smart_domain.association_impl import asociationImplCodeGenerator
from agents.tools.smart_domain.persistent_layer_code_tool import persistentLayerCodeGenerator
from models import llm
class CustomPromptTemplate(StringPromptTemplate):
# The template to use
template: str
# The list of tools available
tools: List[Tool]
def format(self, **kwargs) -> str:
# Get the intermediate steps (AgentAction, Observation tuples)
# Format them in a particular way
intermediate_steps = kwargs.pop("intermediate_steps")
thoughts = ""
for action, observation in intermediate_steps:
thoughts += action.log
thoughts += f"\nObservation: {observation}\nThought: "
# Set the agent_scratchpad variable to that value
kwargs["agent_scratchpad"] = thoughts
# Create a tools variable from the list of tools provided
kwargs["tools"] = "\n".join(
[f"{tool.name}: {tool.description}" for tool in self.tools])
# Create a list of tool names for the tools provided
kwargs["tool_names"] = ", ".join([tool.name for tool in self.tools])
return self.template.format(**kwargs)
class CustomOutputParser(AgentOutputParser):
def parse(self, llm_output: str) -> Union[AgentAction, AgentFinish]:
# Check if agent should finish
if "Final Answer:" in llm_output:
return AgentFinish(
# Return values is generally always a dictionary with a single `output` key
# It is not recommended to try anything else at the moment :)
return_values={"output": llm_output.split(
"Final Answer:")[-1].strip()},
log=llm_output,
)
# Parse out the action and action input
regex = r"Action\s*\d*\s*:(.*?)\nAction\s*\d*\s*Input\s*\d*\s*:[\s]*(.*)"
match = re.search(regex, llm_output, re.DOTALL)
if not match:
raise ValueError(f"Could not parse LLM output: `{llm_output}`")
action = match.group(1).strip()
action_input = match.group(2)
# Return the action and action input
return AgentAction(tool=action, tool_input=action_input.strip(" ").strip('"'), log=llm_output)
# chatllm=ChatOpenAI(temperature=0)
# code_genenrate_memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
# code_generate_agent = initialize_agent(tools, chatllm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, memory=memory, verbose=True)
# agent = initialize_agent(
# tools=tools, llm=llm_chain, template=AGENT_PROMPT, stop=["\nObservation:"], agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, verbose=True)
code_agent_tools = [domainLayerCodeGenerator, entityCodeGenerator, associationCodeGenerator, persistentLayerCodeGenerator, dbEntityRepositoryCodeGenerator, asociationImplCodeGenerator, apiLayerCodeGenerator]
def code_agent_executor() -> AgentExecutor:
output_parser = CustomOutputParser()
AGENT_PROMPT = CustomPromptTemplate(
template=code_generate_agent_template,
tools=code_agent_tools,
# This omits the `agent_scratchpad`, `tools`, and `tool_names` variables because those are generated dynamically
# This includes the `intermediate_steps` variable because that is needed
input_variables=["input", "intermediate_steps"]
)
code_llm_chain = LLMChain(llm=llm(temperature=0.7), prompt=AGENT_PROMPT)
tool_names = [tool.name for tool in code_agent_tools]
code_agent = LLMSingleActionAgent(
llm_chain=code_llm_chain,
output_parser=output_parser,
stop=["\nObservation:"],
allowed_tools=tool_names,
)
code_agent_executor = AgentExecutor.from_agent_and_tools(
agent=code_agent, tools=code_agent_tools, verbose=True)
return code_agent_executor
# if __name__ == "__main__":
# response = domainLayerChain.run("""FeatureConfig用于配置某个Feature中控制前端展示效果的配置项
# FeatureConfig主要属性包括:featureKey(feature标识)、data(配置数据)、saData(埋点数据)、status(状态)、标题、描述、创建时间、更新时间
# FeatureConfig中status为枚举值,取值范围为(DRAFT、PUBLISHED、DISABLED)
# FeatureConfig新增后status为DRAFT、执行发布操作后变为PUBLISHED、执行撤销操作后变为DISABLED
# 状态为DRAFT的FeatureConfig可以执行编辑、发布、撤销操作
# 发布后FeatureConfig变为PUBLISHED状态,可以执行撤销操作
# 撤销后FeatureConfig变为DISABLED状态,不可以执行编辑、发布、撤销操作
# """)
# print(response)
# response = persistentChain.run("""
# Entity:
# ```
# public class FeatureConfig {
# private FeatureConfigId id;
# private FeatureConfigDescription description;
# public enum FeatureConfigStatus {
# DRAFT, PUBLISHED, DISABLED;
# }
# public record FeatureConfigId(String id) {}
# public record FeatureKey(String key) {}
# public record FeatureConfigData(String data) {}
# public record FeatureConfigSaData(String saData) {}
# @Builder
# public record FeatureConfigDescription(FeatureKey featureKey, FeatureConfigData data, FeatureConfigSaData saData, String title, String description,
# FeatureConfigStatus status, LocalDateTime createTime, LocalDateTime updateTime) {}
# public void update(FeatureConfigDescription description) {
# this.title = description.title();
# this.description = description.description();
# this.updateTime = LocalDateTime.now();
# }
# public void publish() {
# this.status = FeatureConfigStatus.PUBLISHED;
# this.updateTime = LocalDateTime.now();
# }
# public void disable() {
# this.status = FeatureConfigStatus.DISABLED;
# this.updateTime = LocalDateTime.now();
# }
# }
# ```
# Association:
# ```
# public interface FeatureConfigs {
# Flux<FeatureConfig> findAllByFeatureKey(String featureKey);
# Mono<FeatureConfig> findById(FeatureConfigId id);
# Mono<FeatureConfig> save(FeatureConfig featureConfig);
# }
# ```
# """)
# print(response)
# response = apiChain.run("""
# Entity:
# ```
# public class FeatureConfig {
# private FeatureConfigId id;
# private FeatureConfigDescription description;
# public enum FeatureConfigStatus {
# DRAFT, PUBLISHED, DISABLED;
# }
# public record FeatureConfigId(String id) {}
# public record FeatureKey(String key) {}
# public record FeatureConfigData(String data) {}
# public record FeatureConfigSaData(String saData) {}
# @Builder
# public record FeatureConfigDescription(FeatureKey featureKey, FeatureConfigData data, FeatureConfigSaData saData, String title, String description,
# FeatureConfigStatus status, LocalDateTime createTime, LocalDateTime updateTime) {}
# public void update(FeatureConfigDescription description) {
# this.title = description.title();
# this.description = description.description();
# this.updateTime = LocalDateTime.now();
# }
# public void publish() {
# this.status = FeatureConfigStatus.PUBLISHED;
# this.updateTime = LocalDateTime.now();
# }
# public void disable() {
# this.status = FeatureConfigStatus.DISABLED;
# this.updateTime = LocalDateTime.now();
# }
# }
# ```
# Association:
# ```
# public interface FeatureConfigs {
# Flux<FeatureConfig> findAllByFeatureKey(String featureKey);
# Mono<FeatureConfig> findById(FeatureConfigId id);
# Mono<FeatureConfig> save(FeatureConfig featureConfig);
# Mono<Void> update(FeatureConfigId id, FeatureConfigDescription description);
# Mono<Void> publish(FeatureConfigId id);
# Mono<Void> disable(FeatureConfigId id);
# }
# ```
# """)
# print(response)
# if __name__ == "code_generate":
# response = code_agent_executor.run("""
# 根据如下需求generate domain layer code:
# ---
# FeatureConfig用于配置某个Feature中控制前端展示效果的配置项
# FeatureConfig主要属性包括:featureKey(feature标识)、data(配置数据)、saData(埋点数据)、status(状态)、标题、描述、创建时间、更新时间
# FeatureConfig中status为枚举值,取值范围为(DRAFT、PUBLISHED、DISABLED)
# FeatureConfig新增后status为DRAFT、执行发布操作后变为PUBLISHED、执行撤销操作后变为DISABLED
# 状态为DRAFT的FeatureConfig可以执行编辑、发布、撤销操作
# 发布后FeatureConfig变为PUBLISHED状态,可以执行撤销操作
# 撤销后FeatureConfig变为DISABLED状态,不可以执行编辑、发布、撤销操作
# ---
# """)
# print(response)