KAG/knext/schema/client.py

199 lines
7.7 KiB
Python
Raw Normal View History

refactor(all): kag v0.6 (#174) * add path find * fix find path * spg guided relation extraction * fix dict parse with same key * rename graphalgoclient to graphclient * rename graphalgoclient to graphclient * file reader supports http url * add checkpointer class * parser supports checkpoint * add build * remove incorrect logs * remove logs * update examples * update chain checkpointer * vectorizer batch size set to 32 * add a zodb backended checkpointer * add a zodb backended checkpointer * fix zodb based checkpointer * add thread for zodb IO * fix(common): resolve mutlithread conflict in zodb IO * fix(common): load existing zodb checkpoints * update examples * update examples * fix zodb writer * add docstring * fix jieba version mismatch * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * 1、fix bug in base_table_splitter * 1、fix bug in base_table_splitter * 1、fix bug in default_chain * 增加solver * add kag * update outline splitter * add main test * add op * code refactor * add tools * fix outline splitter * fix outline prompt * graph api pass * commit with page rank * add search api and graph api * add markdown report * fix vectorizer num batch compute * add retry for vectorize model call * update markdown reader * update markdown reader * update pdf reader * raise extractor failure * add default expr * add log * merge jc reader features * rm import * add build * fix zodb based checkpointer * add thread for zodb IO * fix(common): resolve mutlithread conflict in zodb IO * fix(common): load existing zodb checkpoints * update examples * update examples * fix zodb writer * add docstring * fix jieba version mismatch * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * 1、fix bug in base_table_splitter * 1、fix bug in base_table_splitter * 1、fix bug in default_chain * update outline splitter * add main test * add markdown report * code refactor * fix outline splitter * fix outline prompt * update markdown reader * fix vectorizer num batch compute * add retry for vectorize model call * update markdown reader * raise extractor failure * rm parser * run pipeline * add config option of whether to perform llm config check, default to false * fix * recover pdf reader * several components can be null for default chain * 支持完整qa运行 * add if * remove unused code * 使用chunk兜底 * excluded source relation to choose * add generate * default recall 10 * add local memory * 排除相似边 * 增加保护 * 修复并发问题 * add debug logger * 支持topk参数化 * 支持chunk截断和调整spo select 的prompt * 增加查询请求保护 * 增加force_chunk配置 * fix entity linker algorithm * 增加sub query改写 * fix md reader dup in test * fix * merge knext to kag parallel * fix package * 修复指标下跌问题 * scanner update * scanner update * add doc and update example scripts * fix * add bridge to spg server * add format * fix bridge * update conf for baike * disable ckpt for spg server runner * llm invoke error default raise exceptions * chore(version): bump version to X.Y.Z * update default response generation prompt * add method getSummarizationMetrics * fix(common): fix project conf empty error * fix typo * 增加上报信息 * 修改main solver * postprocessor support spg server * 修改solver支持名 * fix language * 修改chunker接口,增加openapi * rename vectorizer to vectorize_model in spg server config * generate_random_string start with gen * add knext llm vector checker * add knext llm vector checker * add knext llm vector checker * solver移除默认值 * udpate yaml and register_name for baike * udpate yaml and register_name for baike * remove config key check * 修复llmmodule * fix knext project * udpate yaml and register_name for examples * udpate yaml and register_name for examples * Revert "udpate yaml and register_name for examples" This reverts commit b3fa5ca9ba749e501133ac67bd8746027ab839d9. * update register name * fix * fix * support multiple resigter names * update component * update reader register names (#183) * fix markdown reader * fix llm client for retry * feat(common): add processed chunk id checkpoint (#185) * update reader register names * add processed chunk id checkpoint * feat(example): add example config (#186) * update reader register names * add processed chunk id checkpoint * add example config file * add max_workers parameter for getSummarizationMetrics to make it faster * add csqa data generation script generate_data.py * commit generated csqa builder and solver data * add csqa basic project files * adjust split_length and num_threads_per_chain to match lightrag settings * ignore ckpt dirs * add csqa evaluation script eval.py * save evaluation scripts summarization_metrics.py and factual_correctness.py * save LightRAG output csqa_lightrag_answers.json * ignore KAG output csqa_kag_answers.json * add README.md for CSQA * fix(solver): fix solver pipeline conf (#191) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * update links and file paths * reformat csqa kag_config.yaml * reformat csqa python files * reformat getSummarizationMetrics and compare_summarization_answers * fix(solver): fix solver config (#192) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * add except * fix typo in csqa README.md * feat(conf): support reinitialize config for call from java side (#199) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * support reinitialize config for java call * revert default response generation prompt * update project list * add README.md for the hotpotqa, 2wiki and musique examples * 增加spo检索 * turn off kag config dump by default * turn off knext schema dump by default * add .gitignore and fix kag_config.yaml * add README.md for the medicine example * add README.md for the supplychain example * bugfix for risk mining * use exact out * refactor(solver): format solver code (#205) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * support reinitialize config for java call * black format --------- Co-authored-by: peilong <peilong.zpl@antgroup.com> Co-authored-by: 锦呈 <zhangxinhong.zxh@antgroup.com> Co-authored-by: zhengke.gzk <zhengke.gzk@antgroup.com> Co-authored-by: huaidong.xhd <huaidong.xhd@antgroup.com>
2025-01-03 17:10:51 +08:00
# -*- coding: utf-8 -*-
# Copyright 2023 OpenSPG Authors
#
# Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except
# in compliance with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software distributed under the License
# is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express
# or implied.
import os
from typing import List, Dict
import knext.common.cache
from knext.common.base.client import Client
from knext.common.rest import ApiClient, Configuration
from knext.schema import rest
from knext.schema.model.base import BaseSpgType, AlterOperationEnum, SpgTypeEnum
from knext.schema.model.relation import Relation
cache = knext.common.cache.SchemaCache()
CHUNK_TYPE = "Chunk"
2025-06-06 16:26:14 +08:00
TABLE_TYPE = "Table"
feat(kag): update to v0.7 (#456) * add think cost * update csv scanner * add final rerank * add reasoner * add iterative planner * fix dpr search * fix dpr search * add reference data * move odps import * update requirement.txt * update 2wiki * add missing file * fix markdown reader * add iterative planning * update version * update runner * update 2wiki example * update bridge * merge solver and solver_new * add cur day * writer delete * update multi process * add missing files * fix report * add chunk retrieved executor * update try in stream runner result * add path * add math executor * update hotpotqa example * remove log * fix python coder solver * update hotpotqa example * fix python coder solver * update config * fix bad * add log * remove unused code * commit with task thought * move kag model to common * add default chat llm * fix * use static planner * support chunk graph node * add args * support naive rag * llm client support tool calls * add default async * add openai * fix result * fix markdown reader * fix thinker * update asyncio interface * feat(solver): add mcp support (#444) * 上传mcp client相关代码 * 1、完成一套mcp client的调用,从pipeline到planner、executor 2、允许json中传入多个mcp_server,通过大模型进行调用并选择 3、调通baidu_map_mcp的使用 * 1、schema * bugfix:删减冗余代码 --------- Co-authored-by: wanxingyu.wxy <wanxingyu.wxy@antgroup.com> * fix affairqa after solver refactor * fix affairqa after solver refactor * fix readme * add params * update version * update mcp executor * update mcp executor * solver add mcp executor * add missing file * add mpc executor * add executor * x * update * fix requirement * fix main llm config * fix solver * bugfix:修复invoke函数调用逻辑 * chg eva * update example * add kag layer * add step task * support dot refresh * support dot refresh * support dot refresh * support dot refresh * add retrieved num * add retrieved num * add pipelineconf * update ppr * update musique prompts * update * add to_dict for BuilderComponentData * async build * add deduce prompt * add deduce prompt * add deduce prompt * fix reader * add deduce prompt * add page thinker report * modify prmpt * add step status * add self cognition * add self cognition * add memory graph storage * add now time * update memory config * add now time * chg graph loader * 添加prqa数据集和代码 * bugfix:prqa调用逻辑修复 * optimize:优化代码逻辑,生成答案规范化 * add retry py code * update memory graph * update memory graph * fix * fix ner * add with_out_refer generator prompt * fix * close ckpt * fix query * fix query * update version * add llm checker * add llm checker * 1、上传evalutor.py以及修改gold_answer.json格式 2、优化代码逻辑 3、修改README.md文件 * update exp * update exp * rerank support * add static rewrite query * recall more chunks * fix graph load * add static rewrite query * fix bugs * add finish check * add finish check * add finish check * add finish check * 1、上传evalutor.py的结果 2、优化代码逻辑,优化readme文件 * add lf retry * add memory graph api * fix reader api * add ner * add metrics * fix bug * remove ner * add reraise fo retry * add edge prop to memory graph * add memory graph * 1、评测数据集结果修正 2、优化evaluator.py代码 3、删除结果不存在而gold_answer中有答案的问题 * 删除评测结果文件 * fix knext host addr * async eva * add lf prompt * add lf prompt * add config * add retry * add unknown check * add rc result * add rc result * add rc result * add rc result * 依据kag pipeline格式修改代码逻辑并通过测试 * bugfix:删除冗余代码 * fix report prompt * bugfix:触发重试机制 * bugfix:中文符号错误 * fix rethinker prompt * update version to 0.6.2b78 * update version * 1、修改evaluator.py,通过大模型计算准确率,符合最新调用逻辑 2、修改prompt,让没有回答的结果重复测试 * update affairqa for evaluate * update affairqa for evaluate * bugfix:修正数据集 * bugfix:修正数据集 * bugfix:修正数据集 * fix name conflict * bugfix:删除错误问题 * bugfix:文件名命名错误导致evaluator失败 * update for affairqa eval * bugfix:修改代码保持evaluate逻辑一致 * x * update for affairqa readme * remove temp eval scripts * bugfix for math deduce * merge 0.6.2_dev * merge 0.6.2_dev * fix * update client addr * updated version * update for affairqa eval * evaUtils 支持中文 * fix affairqa eval: * remove unused example * update kag config * fix default value * update readme * fix init * 注释信息修改,并添加部分class说明 * update example config * Tc 0.7.0 (#459) * 提交affairQA 代码 * fix affairqa eval --------- Co-authored-by: zhengke.gzk <zhengke.gzk@antgroup.com> * fix all examples * reformat --------- Co-authored-by: peilong <peilong.zpl@antgroup.com> Co-authored-by: 锦呈 <zhangxinhong.zxh@antgroup.com> Co-authored-by: wanxingyu.wxy <wanxingyu.wxy@antgroup.com> Co-authored-by: zhengke.gzk <zhengke.gzk@antgroup.com>
2025-04-17 17:23:52 +08:00
TITLE_TYPE = "Title"
refactor(all): kag v0.6 (#174) * add path find * fix find path * spg guided relation extraction * fix dict parse with same key * rename graphalgoclient to graphclient * rename graphalgoclient to graphclient * file reader supports http url * add checkpointer class * parser supports checkpoint * add build * remove incorrect logs * remove logs * update examples * update chain checkpointer * vectorizer batch size set to 32 * add a zodb backended checkpointer * add a zodb backended checkpointer * fix zodb based checkpointer * add thread for zodb IO * fix(common): resolve mutlithread conflict in zodb IO * fix(common): load existing zodb checkpoints * update examples * update examples * fix zodb writer * add docstring * fix jieba version mismatch * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * 1、fix bug in base_table_splitter * 1、fix bug in base_table_splitter * 1、fix bug in default_chain * 增加solver * add kag * update outline splitter * add main test * add op * code refactor * add tools * fix outline splitter * fix outline prompt * graph api pass * commit with page rank * add search api and graph api * add markdown report * fix vectorizer num batch compute * add retry for vectorize model call * update markdown reader * update markdown reader * update pdf reader * raise extractor failure * add default expr * add log * merge jc reader features * rm import * add build * fix zodb based checkpointer * add thread for zodb IO * fix(common): resolve mutlithread conflict in zodb IO * fix(common): load existing zodb checkpoints * update examples * update examples * fix zodb writer * add docstring * fix jieba version mismatch * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * 1、fix bug in base_table_splitter * 1、fix bug in base_table_splitter * 1、fix bug in default_chain * update outline splitter * add main test * add markdown report * code refactor * fix outline splitter * fix outline prompt * update markdown reader * fix vectorizer num batch compute * add retry for vectorize model call * update markdown reader * raise extractor failure * rm parser * run pipeline * add config option of whether to perform llm config check, default to false * fix * recover pdf reader * several components can be null for default chain * 支持完整qa运行 * add if * remove unused code * 使用chunk兜底 * excluded source relation to choose * add generate * default recall 10 * add local memory * 排除相似边 * 增加保护 * 修复并发问题 * add debug logger * 支持topk参数化 * 支持chunk截断和调整spo select 的prompt * 增加查询请求保护 * 增加force_chunk配置 * fix entity linker algorithm * 增加sub query改写 * fix md reader dup in test * fix * merge knext to kag parallel * fix package * 修复指标下跌问题 * scanner update * scanner update * add doc and update example scripts * fix * add bridge to spg server * add format * fix bridge * update conf for baike * disable ckpt for spg server runner * llm invoke error default raise exceptions * chore(version): bump version to X.Y.Z * update default response generation prompt * add method getSummarizationMetrics * fix(common): fix project conf empty error * fix typo * 增加上报信息 * 修改main solver * postprocessor support spg server * 修改solver支持名 * fix language * 修改chunker接口,增加openapi * rename vectorizer to vectorize_model in spg server config * generate_random_string start with gen * add knext llm vector checker * add knext llm vector checker * add knext llm vector checker * solver移除默认值 * udpate yaml and register_name for baike * udpate yaml and register_name for baike * remove config key check * 修复llmmodule * fix knext project * udpate yaml and register_name for examples * udpate yaml and register_name for examples * Revert "udpate yaml and register_name for examples" This reverts commit b3fa5ca9ba749e501133ac67bd8746027ab839d9. * update register name * fix * fix * support multiple resigter names * update component * update reader register names (#183) * fix markdown reader * fix llm client for retry * feat(common): add processed chunk id checkpoint (#185) * update reader register names * add processed chunk id checkpoint * feat(example): add example config (#186) * update reader register names * add processed chunk id checkpoint * add example config file * add max_workers parameter for getSummarizationMetrics to make it faster * add csqa data generation script generate_data.py * commit generated csqa builder and solver data * add csqa basic project files * adjust split_length and num_threads_per_chain to match lightrag settings * ignore ckpt dirs * add csqa evaluation script eval.py * save evaluation scripts summarization_metrics.py and factual_correctness.py * save LightRAG output csqa_lightrag_answers.json * ignore KAG output csqa_kag_answers.json * add README.md for CSQA * fix(solver): fix solver pipeline conf (#191) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * update links and file paths * reformat csqa kag_config.yaml * reformat csqa python files * reformat getSummarizationMetrics and compare_summarization_answers * fix(solver): fix solver config (#192) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * add except * fix typo in csqa README.md * feat(conf): support reinitialize config for call from java side (#199) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * support reinitialize config for java call * revert default response generation prompt * update project list * add README.md for the hotpotqa, 2wiki and musique examples * 增加spo检索 * turn off kag config dump by default * turn off knext schema dump by default * add .gitignore and fix kag_config.yaml * add README.md for the medicine example * add README.md for the supplychain example * bugfix for risk mining * use exact out * refactor(solver): format solver code (#205) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * support reinitialize config for java call * black format --------- Co-authored-by: peilong <peilong.zpl@antgroup.com> Co-authored-by: 锦呈 <zhangxinhong.zxh@antgroup.com> Co-authored-by: zhengke.gzk <zhengke.gzk@antgroup.com> Co-authored-by: huaidong.xhd <huaidong.xhd@antgroup.com>
2025-01-03 17:10:51 +08:00
OTHER_TYPE = "Others"
TEXT_TYPE = "Text"
INTEGER_TYPE = "Integer"
FLOAT_TYPE = "Float"
BASIC_TYPES = [TEXT_TYPE, INTEGER_TYPE, FLOAT_TYPE]
class SchemaSession:
def __init__(self, client, project_id):
self._alter_spg_types: List[BaseSpgType] = []
self._rest_client = client
self._project_id = project_id
self._spg_types = {}
self.__spg_types = {}
self._init_spg_types()
def _init_spg_types(self):
"""Query project schema and init SPG types in session."""
project_schema = self._rest_client.schema_query_project_schema_get(
self._project_id
)
for spg_type in project_schema.spg_types:
spg_type_name = spg_type.basic_info.name.name
type_class = BaseSpgType.by_type_enum(spg_type.spg_type_enum)
if spg_type.spg_type_enum == SpgTypeEnum.Concept:
self._spg_types[spg_type_name] = type_class(
name=spg_type_name,
hypernym_predicate=spg_type.concept_layer_config.hypernym_predicate,
rest_model=spg_type,
)
else:
self._spg_types[spg_type_name] = type_class(
name=spg_type_name, rest_model=spg_type
)
@property
def spg_types(self) -> Dict[str, BaseSpgType]:
return self._spg_types
def get(self, spg_type_name) -> BaseSpgType:
"""Get SPG type by name from project schema."""
spg_type = self._spg_types.get(spg_type_name)
if spg_type is None:
spg_type = self.__spg_types.get(spg_type_name)
if spg_type is None:
raise ValueError(f"{spg_type_name} is not existed")
else:
return self.__spg_types.get(spg_type_name)
return self._spg_types.get(spg_type_name)
def create_type(self, spg_type: BaseSpgType):
"""Add an SPG type in session with `CREATE` operation."""
spg_type.alter_operation = AlterOperationEnum.Create
self.__spg_types[spg_type.name] = spg_type
self._alter_spg_types.append(spg_type)
return self
def update_type(self, spg_type: BaseSpgType):
"""Add an SPG type in session with `UPDATE` operation."""
spg_type.alter_operation = AlterOperationEnum.Update
self._alter_spg_types.append(spg_type)
return self
def delete_type(self, spg_type: BaseSpgType):
"""Add an SPG type in session with `DELETE` operation."""
spg_type.alter_operation = AlterOperationEnum.Delete
self._alter_spg_types.append(spg_type)
return self
def commit(self):
"""Commit all altered schemas to server."""
schema_draft = []
for spg_type in self._alter_spg_types:
for prop in spg_type.properties.values():
if prop.object_spg_type is None:
object_spg_type = self.get(prop.object_type_name)
prop.object_spg_type = object_spg_type.spg_type_enum
for sub_prop in prop.sub_properties.values():
if sub_prop.object_spg_type is None:
object_spg_type = self.get(sub_prop.object_type_name)
sub_prop.object_spg_type = object_spg_type.spg_type_enum
for rel in spg_type.relations.values():
if rel.is_dynamic is None:
rel.is_dynamic = False
if rel.object_spg_type is None:
object_spg_type = self.get(rel.object_type_name)
rel.object_spg_type = object_spg_type.spg_type_enum
for sub_prop in rel.sub_properties.values():
if sub_prop.object_spg_type is None:
object_spg_type = self.get(sub_prop.object_type_name)
sub_prop.object_spg_type = object_spg_type.spg_type_enum
schema_draft.append(spg_type.to_rest())
if len(schema_draft) == 0:
return
request = rest.SchemaAlterRequest(
project_id=self._project_id, schema_draft=rest.SchemaDraft(schema_draft)
)
key = "KNEXT_DEBUG_DUMP_SCHEMA"
dump_flag = os.getenv(key)
if dump_flag is not None and dump_flag.strip() == "1":
print(request)
else:
print(f"Committing schema: set {key}=1 to dump the schema")
self._rest_client.schema_alter_schema_post(schema_alter_request=request)
class SchemaClient(Client):
""" """
def __init__(self, host_addr: str = None, project_id: str = None):
super().__init__(host_addr, project_id)
self._session = None
self._rest_client: rest.SchemaApi = rest.SchemaApi(
api_client=ApiClient(configuration=Configuration(host=host_addr))
)
def query_spg_type(self, spg_type_name: str) -> BaseSpgType:
"""Query SPG type by name."""
rest_model = self._rest_client.schema_query_spg_type_get(spg_type_name)
type_class = BaseSpgType.by_type_enum(f"{rest_model.spg_type_enum}")
if rest_model.spg_type_enum == SpgTypeEnum.Concept:
return type_class(
name=spg_type_name,
hypernym_predicate=rest_model.concept_layer_config.hypernym_predicate,
rest_model=rest_model,
)
else:
return type_class(name=spg_type_name, rest_model=rest_model)
def query_relation(
self, subject_name: str, predicate_name: str, object_name: str
) -> Relation:
"""Query relation type by s_p_o name."""
rest_model = self._rest_client.schema_query_relation_get(
subject_name, predicate_name, object_name
)
return Relation(
name=predicate_name, object_type_name=object_name, rest_model=rest_model
)
def create_session(self):
"""Create session for altering schema."""
schema_session = cache.get(self._project_id)
if not schema_session:
schema_session = SchemaSession(self._rest_client, self._project_id)
cache.put(self._project_id, schema_session)
return schema_session
def load(self):
schema_session = self.create_session()
schema = {
k.split(".")[-1]: v
for k, v in schema_session.spg_types.items()
if v.spg_type_enum
2025-05-23 14:21:45 +08:00
in [SpgTypeEnum.Concept, SpgTypeEnum.Entity, SpgTypeEnum.Event, SpgTypeEnum.Index]
refactor(all): kag v0.6 (#174) * add path find * fix find path * spg guided relation extraction * fix dict parse with same key * rename graphalgoclient to graphclient * rename graphalgoclient to graphclient * file reader supports http url * add checkpointer class * parser supports checkpoint * add build * remove incorrect logs * remove logs * update examples * update chain checkpointer * vectorizer batch size set to 32 * add a zodb backended checkpointer * add a zodb backended checkpointer * fix zodb based checkpointer * add thread for zodb IO * fix(common): resolve mutlithread conflict in zodb IO * fix(common): load existing zodb checkpoints * update examples * update examples * fix zodb writer * add docstring * fix jieba version mismatch * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * 1、fix bug in base_table_splitter * 1、fix bug in base_table_splitter * 1、fix bug in default_chain * 增加solver * add kag * update outline splitter * add main test * add op * code refactor * add tools * fix outline splitter * fix outline prompt * graph api pass * commit with page rank * add search api and graph api * add markdown report * fix vectorizer num batch compute * add retry for vectorize model call * update markdown reader * update markdown reader * update pdf reader * raise extractor failure * add default expr * add log * merge jc reader features * rm import * add build * fix zodb based checkpointer * add thread for zodb IO * fix(common): resolve mutlithread conflict in zodb IO * fix(common): load existing zodb checkpoints * update examples * update examples * fix zodb writer * add docstring * fix jieba version mismatch * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * commit kag_config-tc.yaml 1、rename type to register_name 2、put a uniqe & specific name to register_name 3、rename reader to scanner 4、rename parser to reader 5、rename num_parallel to num_parallel_file, rename chain_level_num_paralle to num_parallel_chain_of_file 6、rename kag_extractor to schema_free_extractor, schema_base_extractor to schema_constraint_extractor 7、pre-define llm & vectorize_model and refer them in the yaml file Issues to be resolved: 1、examples of event extract & spg extract 2、statistic of indexer, such as nums of nodes & edges extracted, ratio of llm invoke. 3、Exceptions such as Debt, account does not exist should be thrown in llm invoke. 4、conf of solver need to be re-examined. * 1、fix bug in base_table_splitter * 1、fix bug in base_table_splitter * 1、fix bug in default_chain * update outline splitter * add main test * add markdown report * code refactor * fix outline splitter * fix outline prompt * update markdown reader * fix vectorizer num batch compute * add retry for vectorize model call * update markdown reader * raise extractor failure * rm parser * run pipeline * add config option of whether to perform llm config check, default to false * fix * recover pdf reader * several components can be null for default chain * 支持完整qa运行 * add if * remove unused code * 使用chunk兜底 * excluded source relation to choose * add generate * default recall 10 * add local memory * 排除相似边 * 增加保护 * 修复并发问题 * add debug logger * 支持topk参数化 * 支持chunk截断和调整spo select 的prompt * 增加查询请求保护 * 增加force_chunk配置 * fix entity linker algorithm * 增加sub query改写 * fix md reader dup in test * fix * merge knext to kag parallel * fix package * 修复指标下跌问题 * scanner update * scanner update * add doc and update example scripts * fix * add bridge to spg server * add format * fix bridge * update conf for baike * disable ckpt for spg server runner * llm invoke error default raise exceptions * chore(version): bump version to X.Y.Z * update default response generation prompt * add method getSummarizationMetrics * fix(common): fix project conf empty error * fix typo * 增加上报信息 * 修改main solver * postprocessor support spg server * 修改solver支持名 * fix language * 修改chunker接口,增加openapi * rename vectorizer to vectorize_model in spg server config * generate_random_string start with gen * add knext llm vector checker * add knext llm vector checker * add knext llm vector checker * solver移除默认值 * udpate yaml and register_name for baike * udpate yaml and register_name for baike * remove config key check * 修复llmmodule * fix knext project * udpate yaml and register_name for examples * udpate yaml and register_name for examples * Revert "udpate yaml and register_name for examples" This reverts commit b3fa5ca9ba749e501133ac67bd8746027ab839d9. * update register name * fix * fix * support multiple resigter names * update component * update reader register names (#183) * fix markdown reader * fix llm client for retry * feat(common): add processed chunk id checkpoint (#185) * update reader register names * add processed chunk id checkpoint * feat(example): add example config (#186) * update reader register names * add processed chunk id checkpoint * add example config file * add max_workers parameter for getSummarizationMetrics to make it faster * add csqa data generation script generate_data.py * commit generated csqa builder and solver data * add csqa basic project files * adjust split_length and num_threads_per_chain to match lightrag settings * ignore ckpt dirs * add csqa evaluation script eval.py * save evaluation scripts summarization_metrics.py and factual_correctness.py * save LightRAG output csqa_lightrag_answers.json * ignore KAG output csqa_kag_answers.json * add README.md for CSQA * fix(solver): fix solver pipeline conf (#191) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * update links and file paths * reformat csqa kag_config.yaml * reformat csqa python files * reformat getSummarizationMetrics and compare_summarization_answers * fix(solver): fix solver config (#192) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * add except * fix typo in csqa README.md * feat(conf): support reinitialize config for call from java side (#199) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * support reinitialize config for java call * revert default response generation prompt * update project list * add README.md for the hotpotqa, 2wiki and musique examples * 增加spo检索 * turn off kag config dump by default * turn off knext schema dump by default * add .gitignore and fix kag_config.yaml * add README.md for the medicine example * add README.md for the supplychain example * bugfix for risk mining * use exact out * refactor(solver): format solver code (#205) * update reader register names * add processed chunk id checkpoint * add example config file * update solver pipeline config * fix project create * fix main solver conf * support reinitialize config for java call * black format --------- Co-authored-by: peilong <peilong.zpl@antgroup.com> Co-authored-by: 锦呈 <zhangxinhong.zxh@antgroup.com> Co-authored-by: zhengke.gzk <zhengke.gzk@antgroup.com> Co-authored-by: huaidong.xhd <huaidong.xhd@antgroup.com>
2025-01-03 17:10:51 +08:00
}
return schema
def extract_types(self):
schema = self.load()
types = [t for t in schema.keys() if t not in [CHUNK_TYPE] + BASIC_TYPES]
return types
feat(kag): update to v0.7 (#456) * add think cost * update csv scanner * add final rerank * add reasoner * add iterative planner * fix dpr search * fix dpr search * add reference data * move odps import * update requirement.txt * update 2wiki * add missing file * fix markdown reader * add iterative planning * update version * update runner * update 2wiki example * update bridge * merge solver and solver_new * add cur day * writer delete * update multi process * add missing files * fix report * add chunk retrieved executor * update try in stream runner result * add path * add math executor * update hotpotqa example * remove log * fix python coder solver * update hotpotqa example * fix python coder solver * update config * fix bad * add log * remove unused code * commit with task thought * move kag model to common * add default chat llm * fix * use static planner * support chunk graph node * add args * support naive rag * llm client support tool calls * add default async * add openai * fix result * fix markdown reader * fix thinker * update asyncio interface * feat(solver): add mcp support (#444) * 上传mcp client相关代码 * 1、完成一套mcp client的调用,从pipeline到planner、executor 2、允许json中传入多个mcp_server,通过大模型进行调用并选择 3、调通baidu_map_mcp的使用 * 1、schema * bugfix:删减冗余代码 --------- Co-authored-by: wanxingyu.wxy <wanxingyu.wxy@antgroup.com> * fix affairqa after solver refactor * fix affairqa after solver refactor * fix readme * add params * update version * update mcp executor * update mcp executor * solver add mcp executor * add missing file * add mpc executor * add executor * x * update * fix requirement * fix main llm config * fix solver * bugfix:修复invoke函数调用逻辑 * chg eva * update example * add kag layer * add step task * support dot refresh * support dot refresh * support dot refresh * support dot refresh * add retrieved num * add retrieved num * add pipelineconf * update ppr * update musique prompts * update * add to_dict for BuilderComponentData * async build * add deduce prompt * add deduce prompt * add deduce prompt * fix reader * add deduce prompt * add page thinker report * modify prmpt * add step status * add self cognition * add self cognition * add memory graph storage * add now time * update memory config * add now time * chg graph loader * 添加prqa数据集和代码 * bugfix:prqa调用逻辑修复 * optimize:优化代码逻辑,生成答案规范化 * add retry py code * update memory graph * update memory graph * fix * fix ner * add with_out_refer generator prompt * fix * close ckpt * fix query * fix query * update version * add llm checker * add llm checker * 1、上传evalutor.py以及修改gold_answer.json格式 2、优化代码逻辑 3、修改README.md文件 * update exp * update exp * rerank support * add static rewrite query * recall more chunks * fix graph load * add static rewrite query * fix bugs * add finish check * add finish check * add finish check * add finish check * 1、上传evalutor.py的结果 2、优化代码逻辑,优化readme文件 * add lf retry * add memory graph api * fix reader api * add ner * add metrics * fix bug * remove ner * add reraise fo retry * add edge prop to memory graph * add memory graph * 1、评测数据集结果修正 2、优化evaluator.py代码 3、删除结果不存在而gold_answer中有答案的问题 * 删除评测结果文件 * fix knext host addr * async eva * add lf prompt * add lf prompt * add config * add retry * add unknown check * add rc result * add rc result * add rc result * add rc result * 依据kag pipeline格式修改代码逻辑并通过测试 * bugfix:删除冗余代码 * fix report prompt * bugfix:触发重试机制 * bugfix:中文符号错误 * fix rethinker prompt * update version to 0.6.2b78 * update version * 1、修改evaluator.py,通过大模型计算准确率,符合最新调用逻辑 2、修改prompt,让没有回答的结果重复测试 * update affairqa for evaluate * update affairqa for evaluate * bugfix:修正数据集 * bugfix:修正数据集 * bugfix:修正数据集 * fix name conflict * bugfix:删除错误问题 * bugfix:文件名命名错误导致evaluator失败 * update for affairqa eval * bugfix:修改代码保持evaluate逻辑一致 * x * update for affairqa readme * remove temp eval scripts * bugfix for math deduce * merge 0.6.2_dev * merge 0.6.2_dev * fix * update client addr * updated version * update for affairqa eval * evaUtils 支持中文 * fix affairqa eval: * remove unused example * update kag config * fix default value * update readme * fix init * 注释信息修改,并添加部分class说明 * update example config * Tc 0.7.0 (#459) * 提交affairQA 代码 * fix affairqa eval --------- Co-authored-by: zhengke.gzk <zhengke.gzk@antgroup.com> * fix all examples * reformat --------- Co-authored-by: peilong <peilong.zpl@antgroup.com> Co-authored-by: 锦呈 <zhangxinhong.zxh@antgroup.com> Co-authored-by: wanxingyu.wxy <wanxingyu.wxy@antgroup.com> Co-authored-by: zhengke.gzk <zhengke.gzk@antgroup.com>
2025-04-17 17:23:52 +08:00
def extract_types_zh_mapping(self):
schema = self.load()
schema_mapping = {
t: schema[t].name_zh for t in schema.keys() if t not in [CHUNK_TYPE] + BASIC_TYPES
}
return schema_mapping