MCPcopy Create free account

hub / github.com/DemonDamon/FinnewsHunter / functions

Functions1,198 in github.com/DemonDamon/FinnewsHunter

⨍ Functions 1,198◇ Types & classes 338↳ Endpoints 72
↓ 2 callersMethodextractData
Extract column data with tag in 'tag_list' to the list.
legacy_v1/Crawler/crawler_sina.py:278
↓ 2 callersMethodextractData
Extract column data with tag in 'tag_list' to the list.
legacy_v1/Crawler/crawler_jrj.py:298
↓ 2 callersMethodextract_search_requests
从文本中提取搜索请求 支持格式: - [SEARCH: "query"] - [SEARCH: "query" source:akshare] - [SEARCH: "query" source:bo
backend/app/agents/search_analyst.py:79
↓ 2 callersMethodforward
前向传播 Args: tokens: 输入 token 序列 [batch, seq_len] Returns: logits: 下一个 token 的 lo
backend/app/alpha_mining/model/alpha_generator.py:104
↓ 2 callersMethodgenerate_search_plan
生成搜索计划（不执行） 根据用户问题分析需要哪些数据，生成待确认的搜索计划 Args: query: 用户问题 stock_code: 股票代码
backend/app/agents/data_collector_v2.py:155
↓ 2 callersMethodgenerate_search_queries
生成多样化的搜索查询组合 Args: max_queries: 最大查询数量 Returns: 查询字符串列表
backend/app/knowledge/graph_models.py:146
↓ 2 callersFunctiongetCategoryConfig
(t: any)
frontend/src/components/alpha-mining/OperatorGrid.tsx:104
↓ 2 callersFunctiongetDebateModes
(t: any)
frontend/src/components/DebateConfig.tsx:38
↓ 2 callersFunctiongetModeIcon
(mode: DebateMode)
frontend/src/components/DebateConfig.tsx:117
↓ 2 callersFunctiongetRoleConfig
(t: any)
frontend/src/components/DebateChatRoom.tsx:52
↓ 2 callersFunctionget_crawler
获取全局爬虫实例
backend/app/tools/crawler_enhanced.py:802
↓ 2 callersFunctionget_embedding_service
获取 Embedding 服务实例（单例模式） Returns: EmbeddingService 实例
backend/app/services/embedding_service.py:301
↓ 2 callersMethodget_reward
获取强化学习奖励 使用 Sortino Ratio 作为奖励信号。 Args: factor: 因子值 returns: 收益率
backend/app/alpha_mining/backtest/evaluator.py:235
↓ 2 callersFunctionget_stock_count
获取数据库中股票数量
backend/app/scripts/init_stocks.py:267
↓ 2 callersMethodhealth_check
健康检查
backend/app/core/neo4j_client.py:118
↓ 2 callersMethodinfo
(self)
backend/app/financial/providers/sina/provider.py:19
↓ 2 callersMethodinfo
(self)
backend/app/financial/providers/yicai/provider.py:19
↓ 2 callersMethodinfo
(self)
backend/app/financial/providers/eastmoney/provider.py:19
↓ 2 callersMethodinfo
(self)
backend/app/financial/providers/nbd/provider.py:19
↓ 2 callersMethodinfo
(self)
backend/app/financial/providers/tencent/provider.py:19
↓ 2 callersMethodinfo
(self)
backend/app/financial/providers/netease/provider.py:19
↓ 2 callersMethodlist_all_companies
列出所有公司
backend/app/knowledge/graph_service.py:564
↓ 2 callersMethodrenew_dictionary
(self, old_dict_path, new_raw_documents_list,
legacy_v1/src/Leorio/topicmodelling.py:65
↓ 2 callersMethodsearch_on_360
在 360 搜索上搜索并获取结果 Args: query: 搜索关键词 num_results: 获取的结果数量 Returns: 搜
backend/app/tools/interactive_crawler.py:516
↓ 2 callersMethodsearch_on_baidu
在百度上搜索并获取结果（百度对简单爬虫相对友好） Args: query: 搜索关键词 num_results: 获取的结果数量 Returns:
backend/app/tools/interactive_crawler.py:193
↓ 2 callersMethodsearch_on_baidu_news
在百度新闻搜索（news.baidu.com）获取新闻结果 使用 news.baidu.com 入口，返回的 URL 是真实的第三方新闻链接， 不是百度跳转链接，避免乱码问题。 Args:
backend/app/tools/interactive_crawler.py:302
↓ 2 callersMethodsearch_on_sogou
在搜狗上搜索并获取结果（备用搜索引擎） Args: query: 搜索关键词 num_results: 获取的结果数量 Returns:
backend/app/tools/interactive_crawler.py:420
↓ 2 callersMethodtransform_query
将标准参数转换为爬虫参数 Args: params: 标准查询参数 Returns: 爬虫参数字典
backend/app/financial/providers/sina/fetchers/news.py:82
↓ 2 callersMethodtransform_vectorized_corpus
(corpora_dictionary, bow_vector, model
legacy_v1/src/Leorio/topicmodelling.py:113
↓ 2 callersMethodupdate_from_news
根据新闻更新图谱 Args: stock_code: 股票代码 news_content: 新闻内容 extracted_info: 提取的信息（由 LLM 提取）
backend/app/knowledge/graph_service.py:457
↓ 2 callersFunctionuseNewsToolbar
()
frontend/src/context/NewsToolbarContext.tsx:32
↓ 2 callersFunctionvectorize_all_pending
向量化所有未向量化但已分析的新闻
backend/tests/manual_vectorize.py:120
↓ 2 callersFunctionvectorize_news_manually
手动向量化单个新闻
backend/tests/manual_vectorize.py:22
↓ 1 callersMethodConnDB
Connect to the mongodb.
legacy_v1/Text_Analysis/text_mining.py:48
↓ 1 callersMethodConnDB
Connect mongodb.
legacy_v1/Crawler/crawler_nbd.py:244
↓ 1 callersMethodConnDB
(self)
legacy_v1/Crawler/crawler_tushare.py:15
↓ 1 callersMethodCrawlCompanyNews
Crawl historical company news
legacy_v1/Crawler/crawler_nbd.py:179
↓ 1 callersMethodCrawlHistoryCompanyNews
Crawl historical company news
legacy_v1/Crawler/crawler_sina.py:206
↓ 1 callersMethodCrawlRealtimeCompanyNews
Continue crawling company news from first website page every once in a while and extract the useful information, including su
legacy_v1/Crawler/crawler_stcn.py:102
↓ 1 callersMethodCrawlRealtimeCompanyNews
Continue crawling company news from first website page every once in a while and extract the useful information, including sum
legacy_v1/Crawler/crawler_cnstock.py:164
↓ 1 callersMethodCrawlRealtimeCompanyNews
Continue crawling company news from first website page every once in a while and extract the useful information, including su
legacy_v1/Crawler/crawler_sina.py:143
↓ 1 callersMethodCrawlRealtimeCompanyNews
Continue crawling company news from first website page every once in a while and extract the useful information, including su
legacy_v1/Crawler/crawler_jrj.py:149
↓ 1 callersMethodRdForestClassifier
Random Forest Classifier. # Arguments: train_X: Features train data. train_Y: Labels train data. test_X: Features train data. te
legacy_v1/Text_Analysis/text_mining.py:424
↓ 1 callersMethodReCrawlArticles
Continue crawling urls without main information return. # Arguments: url_list: List of urls without getting any articles(main bo
legacy_v1/Crawler/crawler_nbd.py:150
↓ 1 callersMethodReCrawlNews
Continue crawling pages without any return. # Arguments: url_list: List of web pages that without any values.
legacy_v1/Crawler/crawler_nbd.py:107
↓ 1 callersMethodSVMClassifier
SVM Classifier. # Arguments: train_X: Features train data. train_Y: Labels train data. test_X: Features train data. test_Y: Labe
legacy_v1/Text_Analysis/text_mining.py:386
↓ 1 callersMethod__init__
(self, extractor_agent: KnowledgeExtractorAgent)
backend/app/knowledge/knowledge_extractor.py:678
↓ 1 callersMethod__init__
(self)
backend/app/financial/tools.py:149
↓ 1 callersMethod__init__
(self, llm_provider=None, organization_id: str = "finnews")
backend/app/agents/data_collector_v2.py:730
↓ 1 callersMethod__init__
(self, llm_provider=None, organization_id: str = "finnews")
backend/app/agents/data_collector.py:260
↓ 1 callersMethod_action_evaluate
评估因子表达式
backend/app/alpha_mining/tools/alpha_mining_tool.py:277
↓ 1 callersMethod_action_generate
生成候选因子
backend/app/alpha_mining/tools/alpha_mining_tool.py:316
↓ 1 callersMethod_action_list
列出已发现的因子
backend/app/alpha_mining/tools/alpha_mining_tool.py:354
↓ 1 callersMethod_action_mine
执行因子挖掘
backend/app/alpha_mining/tools/alpha_mining_tool.py:226
↓ 1 callersMethod_add_industry
添加行业节点（内部方法）
backend/app/knowledge/graph_service.py:324
↓ 1 callersMethod_analyze_with_llm
使用 LLM 分析需要哪些数据
backend/app/agents/data_collector_v2.py:280
↓ 1 callersMethod_apply_decay
应用指数衰减平滑 情感影响会随时间衰减，使用指数移动平均来平滑
backend/app/alpha_mining/features/sentiment.py:177
↓ 1 callersMethod_build_from_dataframe
从 DataFrame 构建特征 支持两种格式： 1. 单股票：index=date, columns=[close, volume, ...] 2. 多股票：MultiIndex 或 pivot 后的 DataFr
backend/app/alpha_mining/features/market.py:84
↓ 1 callersMethod_build_from_tensors
从张量字典构建特征 Args: data: 包含 close, volume 等张量的字典 每个张量形状为 [batch, time_steps] 或 [time_steps]
backend/app/alpha_mining/features/market.py:142
↓ 1 callersMethod_build_search_query
构建搜索查询字符串（添加时间限制） Args: query: 原始查询 days: 时间范围 Returns: 增强的搜索查询
backend/app/tools/search_engine_crawler.py:122
↓ 1 callersMethod_calc_ic
计算 IC (Information Coefficient) 因子值与下期收益的 Pearson 相关系数
backend/app/alpha_mining/backtest/evaluator.py:194
↓ 1 callersMethod_calc_max_drawdown
计算最大回撤
backend/app/alpha_mining/backtest/evaluator.py:226
↓ 1 callersMethod_calc_rank_ic
计算 Rank IC 因子排名与收益排名的 Spearman 相关系数
backend/app/alpha_mining/backtest/evaluator.py:209
↓ 1 callersMethod_calc_sortino
计算 Sortino Ratio 只考虑下行风险（负收益的标准差）
backend/app/alpha_mining/backtest/evaluator.py:169
↓ 1 callersMethod_check_manager_interrupt
检查投资经理是否要打断辩论
backend/app/agents/orchestrator.py:797
↓ 1 callersMethod_check_manager_interrupt_or_search
检查投资经理是否要打断辩论或请求更多数据 Returns: (should_interrupt: bool, additional_data: str or None)
backend/app/agents/orchestrator.py:822
↓ 1 callersMethod_clean_text
清理文本
backend/app/financial/providers/yicai/fetchers/news.py:249
↓ 1 callersMethod_clean_text
清理文本
backend/app/financial/providers/eastmoney/fetchers/news.py:257
↓ 1 callersMethod_clean_text
清理文本
backend/app/financial/providers/nbd/fetchers/news.py:256
↓ 1 callersMethod_clean_text
清理文本
backend/app/financial/providers/tencent/fetchers/news.py:288
↓ 1 callersMethod_clean_text
清理文本
backend/app/financial/providers/netease/fetchers/news.py:255
↓ 1 callersMethod_close_driver
关闭 WebDriver
backend/app/tools/dynamic_crawler_example.py:56
↓ 1 callersMethod_collect_financial
搜集财务数据
backend/app/agents/data_collector.py:124
↓ 1 callersMethod_collect_fund_flow
搜集资金流向数据
backend/app/agents/data_collector.py:134
↓ 1 callersMethod_collect_news
搜集新闻数据
backend/app/agents/data_collector.py:103
↓ 1 callersMethod_collect_realtime_quote
搜集实时行情
backend/app/agents/data_collector.py:144
↓ 1 callersMethod_crawl_news_detail
爬取新闻详情页 Args: url: 新闻URL Returns: 新闻项或None
backend/app/tools/sina_crawler.py:108
↓ 1 callersMethod_crawl_news_detail
爬取新闻详情
backend/app/financial/providers/sina/fetchers/news.py:211
↓ 1 callersMethod_crawl_page
爬取单页新闻
backend/app/tools/yicai_crawler.py:56
↓ 1 callersMethod_crawl_page
爬取单页新闻
backend/app/tools/jingji21_crawler.py:57
↓ 1 callersMethod_crawl_page
爬取单页新闻
backend/app/tools/jwview_crawler.py:57
↓ 1 callersMethod_crawl_page
爬取单页新闻
backend/app/tools/caijing_crawler.py:57
↓ 1 callersMethod_crawl_page
爬取单页新闻
backend/app/tools/eastmoney_crawler.py:56
↓ 1 callersMethod_crawl_page
爬取单页新闻列表 Args: page: 页码（目前只支持首页，忽略此参数） Returns: 新闻列表
backend/app/tools/sina_crawler.py:57
↓ 1 callersMethod_crawl_page
爬取单页新闻
backend/app/tools/netease163_crawler.py:56
↓ 1 callersMethod_crawl_page
爬取单页新闻
backend/app/tools/nbd_crawler.py:56
↓ 1 callersMethod_crawl_page
爬取单页新闻（使用API） Args: page: 页码 Returns: 新闻列表
backend/app/tools/eeo_crawler.py:158
↓ 1 callersMethod_crawl_page
爬取单页新闻 优先使用API获取新闻，如果API失败则回退到HTML解析 Args: page: 页码 Returns: 新
backend/app/tools/tencent_crawler.py:59
↓ 1 callersMethod_crawl_page
爬取单个页面
backend/app/financial/providers/sina/fetchers/news.py:144
↓ 1 callersMethod_crawl_page_html
备用方案：直接解析HTML页面（只能获取首屏内容）
backend/app/tools/eeo_crawler.py:267
↓ 1 callersMethod_crawl_with_selenium
使用 Selenium 爬取动态加载的内容
backend/app/tools/dynamic_crawler_example.py:89
↓ 1 callersMethod_create_provider
创建 LLM 提供者
backend/app/services/llm_service.py:84
↓ 1 callersMethod_create_task
创建搜索任务
backend/app/agents/data_collector_v2.py:238
↓ 1 callersMethod_ensure_constraints
确保数据库约束和索引存在
backend/app/knowledge/graph_service.py:34
↓ 1 callersMethod_evaluate_single
评估单个样本
backend/app/alpha_mining/backtest/evaluator.py:101
↓ 1 callersMethod_execute_task
执行单个搜索任务
backend/app/agents/data_collector_v2.py:432
↓ 1 callersMethod_extract_author
提取作者
backend/app/tools/yicai_crawler.py:202
↓ 1 callersMethod_extract_author
提取作者
backend/app/tools/jingji21_crawler.py:218
↓ 1 callersMethod_extract_author
提取作者
backend/app/tools/jwview_crawler.py:219
↓ 1 callersMethod_extract_author
提取作者
backend/app/tools/caijing_crawler.py:248
← previousnext →201–300 of 1,198, ranked by callers