diff --git a/.idea/.gitignore b/.idea/.gitignore new file mode 100644 index 0000000..35410ca --- /dev/null +++ b/.idea/.gitignore @@ -0,0 +1,8 @@ +# 默认忽略的文件 +/shelf/ +/workspace.xml +# 基于编辑器的 HTTP 客户端请求 +/httpRequests/ +# Datasource local storage ignored files +/dataSources/ +/dataSources.local.xml diff --git a/.idea/misc.xml b/.idea/misc.xml new file mode 100644 index 0000000..0548357 --- /dev/null +++ b/.idea/misc.xml @@ -0,0 +1,6 @@ + + + + + + \ No newline at end of file diff --git a/.idea/modules.xml b/.idea/modules.xml new file mode 100644 index 0000000..67b4c1d --- /dev/null +++ b/.idea/modules.xml @@ -0,0 +1,8 @@ + + + + + + + + \ No newline at end of file diff --git a/.idea/民航NLP.iml b/.idea/民航NLP.iml new file mode 100644 index 0000000..d6ebd48 --- /dev/null +++ b/.idea/民航NLP.iml @@ -0,0 +1,9 @@ + + + + + + + + + \ No newline at end of file diff --git a/examples/EE-eng.jpg b/examples/EE-eng.jpg new file mode 100644 index 0000000..789205c Binary files /dev/null and b/examples/EE-eng.jpg differ diff --git a/examples/EE-zh.jpg b/examples/EE-zh.jpg new file mode 100644 index 0000000..12d8965 Binary files /dev/null and b/examples/EE-zh.jpg differ diff --git a/examples/NER-eng.jpg b/examples/NER-eng.jpg new file mode 100644 index 0000000..52e6f8c Binary files /dev/null and b/examples/NER-eng.jpg differ diff --git a/examples/NER-zh.jpg b/examples/NER-zh.jpg new file mode 100644 index 0000000..bf9635e Binary files /dev/null and b/examples/NER-zh.jpg differ diff --git a/examples/RE-eng.jpg b/examples/RE-eng.jpg new file mode 100644 index 0000000..93015df Binary files /dev/null and b/examples/RE-eng.jpg differ diff --git a/examples/RE-zh.jpg b/examples/RE-zh.jpg new file mode 100644 index 0000000..ee3f4c2 Binary files /dev/null and b/examples/RE-zh.jpg differ diff --git a/examples/readme.md b/examples/readme.md new file mode 100644 index 0000000..f158fc0 --- /dev/null +++ b/examples/readme.md @@ -0,0 +1 @@ +# some run examples diff --git a/moxun b/moxun new file mode 160000 index 0000000..cd0bdc0 --- /dev/null +++ b/moxun @@ -0,0 +1 @@ +Subproject commit cd0bdc08708d2d7cd7f278abf44801bb431d739f diff --git a/public/favicon.ico b/public/favicon.ico new file mode 100644 index 0000000..9941ace Binary files /dev/null and b/public/favicon.ico differ diff --git a/public/index.html b/public/index.html new file mode 100644 index 0000000..8ba0672 --- /dev/null +++ b/public/index.html @@ -0,0 +1,43 @@ + + + + + + + + + + + + + + 基于有噪声民航数据的NLP处理系统 + + + + +
+ + + + \ No newline at end of file diff --git a/public/logo192.png b/public/logo192.png new file mode 100644 index 0000000..9941ace Binary files /dev/null and b/public/logo192.png differ diff --git a/public/logo512.png b/public/logo512.png new file mode 100644 index 0000000..9941ace Binary files /dev/null and b/public/logo512.png differ diff --git a/public/manifest.json b/public/manifest.json new file mode 100644 index 0000000..2d130cd --- /dev/null +++ b/public/manifest.json @@ -0,0 +1,25 @@ +{ + "short_name": "GPT4IE", + "name": "GPT4IE", + "icons": [ + { + "src": "favicon.ico", + "sizes": "64x64 32x32 24x24 16x16", + "type": "image/x-icon" + }, + { + "src": "logo192.png", + "type": "image/png", + "sizes": "192x192" + }, + { + "src": "logo512.png", + "type": "image/png", + "sizes": "512x512" + } + ], + "start_url": ".", + "display": "standalone", + "theme_color": "#000000", + "background_color": "#ffffff" +} \ No newline at end of file diff --git a/public/prompts/EE-chi.prompt b/public/prompts/EE-chi.prompt new file mode 100644 index 0000000..092d9af --- /dev/null +++ b/public/prompts/EE-chi.prompt @@ -0,0 +1,4 @@ +已知事件类型和论元角色列表的对应关系如下：$prompt1 + +给定一句话："$prompt", 请提取出这句话的事件论元，并以字典的形式回答，字典的形式为：{事件类型: {论元角色: 论元内容}}。直接给出答案即可。如果没有论元角色没有相应的论元内容，论元内容回答：无。 + diff --git a/public/prompts/EE-eng.prompt b/public/prompts/EE-eng.prompt new file mode 100644 index 0000000..2a4faae --- /dev/null +++ b/public/prompts/EE-eng.prompt @@ -0,0 +1,3 @@ +The correspondence between known event types and argument role lists is as follows: $prompt1 + +Give a sentence:"$prompt", Please extract the event argument or arguments of this sentence, and answer it or them in the form of a dictionary. The dictionary is in the form of: {event type: {argument role: argument content}}. Just give the answer. If the argument role does not have the corresponding argument content, the argument content answers: None. \ No newline at end of file diff --git a/public/prompts/NER-chi.prompt b/public/prompts/NER-chi.prompt new file mode 100644 index 0000000..86485b2 --- /dev/null +++ b/public/prompts/NER-chi.prompt @@ -0,0 +1,5 @@ +已知实体类型列表：$prompt1。 + +给定一个句子："$prompt" + +在给定的句子中可能包含哪些实体，请按照形式 (实体类型, 实体内容) 回答： \ No newline at end of file diff --git a/public/prompts/NER-eng.prompt b/public/prompts/NER-eng.prompt new file mode 100644 index 0000000..cc5997c --- /dev/null +++ b/public/prompts/NER-eng.prompt @@ -0,0 +1,5 @@ +List of known entity types: $prompt1. + +Given a sentence: "$prompt" + +Which entities may be included in a given sentence, please answer in the form (entity type, entity content): \ No newline at end of file diff --git a/public/prompts/RE-chi.prompt b/public/prompts/RE-chi.prompt new file mode 100644 index 0000000..1e731c9 --- /dev/null +++ b/public/prompts/RE-chi.prompt @@ -0,0 +1,11 @@ +假设你是一个实体关系三元组抽取模型。我会给你头实体类型列表subject_types，尾实体类型列表object_types，关系列表relations，再给你一个句子，请你根据这三个列表抽出句子中的subject和object，并组成三元组，且形式为(subject, relation, object)。 + +给定的句子为："$prompt" + +relations：$prompt1 + +subject_types：$prompt2 + +object_types：$prompt3 + +在给定的句子中，可能包含了哪些三元组？请按照形式(subject, relation, object)回答： \ No newline at end of file diff --git a/public/prompts/RE-eng.prompt b/public/prompts/RE-eng.prompt new file mode 100644 index 0000000..dd35c30 --- /dev/null +++ b/public/prompts/RE-eng.prompt @@ -0,0 +1,11 @@ +Suppose you are an entity-relationship triple extraction model. I'll give you list of head entity types: subject_types, list of tail entity types: object_types, list of relations: relations. Give you a sentence, please extract the subject and object in the sentence based on these three lists, and form a triplet in the form of (subject, relation, object). + +The given sentence is "$prompt". + +relations: $prompt1 + +subject_types: $prompt2 + +object_types: $prompt3 + +In the given sentence, what triples might be contained? Please answer in the form (subject, relation, object): \ No newline at end of file diff --git a/public/prompts/temp.txt b/public/prompts/temp.txt new file mode 100644 index 0000000..63814b2 --- /dev/null +++ b/public/prompts/temp.txt @@ -0,0 +1,38 @@ +The aid group Doctors Without Borders said that since Saturday , more than 275 wounded people had been admitted and treated at Donka Hospital in the capital of Guinea , Conakry . + +['location-located_in', 'administrative_division-country', 'person-place_lived', 'person-company', 'person-nationality', 'company-founders', 'country-administrative_divisions', 'person-children', 'country-capital', 'deceased_person-place_of_death', 'neighborhood-neighborhood_of', 'person-place_of_birth'] + +['organization', 'person', 'location', 'country'] + +['person', 'location', 'country', 'organization', 'city'] + +_____________________________________________________ + + +第五部：《如懿传》《如懿传》是一部古装宫廷情感电视剧，由汪俊执导，周迅、霍建华、张钧甯、董洁、辛芷蕾、童瑶、李纯、邬君梅等主演 + +['所属专辑', '成立日期', '海拔', '官方语言', '占地面积', '父亲', '歌手', '制片人', '导演', '首都', '主演', '董事长', '祖籍', '妻子', '母亲', '气候', '面积', '主角', '邮政编码', '简称', '出品公司', '注册资本', '编剧', '创始人', '毕业院校', '国籍', '专业代码', '朝代', '作者', '作词', '所在城市', '嘉宾', '总部地点', '人口数量', '代言人', '改编自', '校长', '丈夫', '主持人', '主题曲', '修业年限', '作曲', '号', '上映时间', '票房', '饰演', '配音', '获奖'] + +['国家', '行政区', '文学作品', '人物', '影视作品', '学校', '图书作品', '地点', '历史人物', '景点', '歌曲', '学科专业', '企业', '电视综艺', '机构', '企业/品牌', '娱乐人物'] + +['国家', '人物', 'Text', 'Date', '地点', '气候', '城市', '歌曲', '企业', 'Number', '音乐专辑', '学校', '作品', '语言'] +_______________________________________________________ +比如本轮的比赛中就有，中甲榜首之争，再次惨败的西北狼回到主场能否止住颓势，闹剧频出的川足事态如何发展，永昌客场是否能迎来首胜。 + +{'组织行为-罢工': ['时间', '所属组织', '罢工人数', '罢工人员'], '竞赛行为-晋级': ['时间', '晋级方', '晋级赛事'], '财经/交易-涨停':['时间', '涨停股票'] , '组织关系-解雇': ['时间', '解雇方', '被解雇人员']} +_____________________________________________________ +He lost an election to a dead man. + +{'Personnel:Elect': ['Person', 'Entity', 'Position', 'Time', 'Place'], 'Business:Declare-Bankruptcy': ['Org', 'Time', 'Place'], 'Justice:Arrest-Jail': ['Person', 'Agent', 'Crime', 'Time', 'Place'], 'Life:Divorce': ['Person', 'Time', 'Place'], 'Life:Injure': ['Agent', 'Victim', 'Instrument', 'Time', 'Place']} +———————————————————————————————————————————— +在过去的五年中，致公党在邓小平理论指引下，遵循社会主义初级阶段的基本路线，努力实践致公党十大提出的发挥参政党职能、加强自身建设的基本任务。 + +['组织机构', '地点', '人物'] + +_____________________________________________________ +Japan then laid siege to the Syrian penalty area and had a goal disallowed for offside in the 16th minute. + +['LOC', 'MISC', 'ORG', 'PER'] + +_____________________________________________________ +"homepage": "https://cocacola-lab.github.io/GPT4IE", \ No newline at end of file diff --git a/public/robots.txt b/public/robots.txt new file mode 100644 index 0000000..e9e57dc --- /dev/null +++ b/public/robots.txt @@ -0,0 +1,3 @@ +# https://www.robotstxt.org/robotstxt.html +User-agent: * +Disallow: diff --git a/src/App.css b/src/App.css new file mode 100644 index 0000000..317bd6a --- /dev/null +++ b/src/App.css @@ -0,0 +1,210 @@ +@import url('https://fonts.googleapis.com/css2?family=Noticia+Text:wght@400;700&display=swap'); + +.headerText { + font-family: 'Noticia Text'; + font-weight: 900; + font-size: 60px; + text-align: center; + padding-top: 40px; +} + +.subheaderText { + font-family: 'Noticia Text'; + font-weight: 400; + font-size: 20px; + text-align: center; + width: 60%; + margin: 0 auto; +} + +.subsubheaderText { + font-family: 'Noticia Text'; + font-weight: 400; + font-size: 18px; + text-align: center; + width: 60%; + margin: 0 auto; +} + + +table{ + font-family: 'Oswald', sans-serif; + border-collapse:collapse; + margin-left:auto; + margin-right:auto; + margin-top: 5px; +} + +td{ + background-color:#ffffff; + width:50%; + height:50px; + text-align:center; +} + +tr{ + border-bottom: 1px solid #dddddd; +} + +tr:first-of-type{ + border-top: 2px solid gray; +} + +tr:last-of-type{ + border-bottom: 2px solid gray; +} + + +.opensourceText { + font-family: 'Noticia Text'; + font-weight: 400; + font-size: 14px; + text-align: center; + width: 60%; + margin: 10px auto; +} + +.container { + padding-right: 15px; + padding-left: 15px; + margin-right: auto; + margin-left: auto; +} + +@media (min-width: 768px) { + .container { + width: 750px; + } +} + +@media (min-width: 992px) { + .container { + width: 970px; + } +} + +@media (min-width: 1200px) { + .container { + width: 1170px; + } +} + +.graphContainer { + width: 80%; + height: 80%; + background-color: rgb(255, 255, 255); + border-radius: 10px; + overflow: hidden; + border: 1px solid rgb(0, 0, 0); + box-shadow: 0 0 5px rgb(0 0 0 / 10%); + margin: 0 auto; + margin-bottom: 10px; +} + +.ulC { + text-align: center +} + +.footer { + font-family: 'Noticia Text'; + font-weight: 400; + font-size: 13px; + text-align: center; + margin: 0 auto; + margin-bottom: 100px; +} + +.inputContainer { + width: 60%; + margin: 0 auto; + margin-top: 50px; + margin-bottom: 50px; +} + +.searchBar { + border: 1px solid rgb(0, 0, 0); + border-radius: 5px; + box-shadow: 0 0 5px rgb(0 0 0 / 10%); + display: block; + font-size: 16px; + height: 50px; + outline: none; + padding: 0 10px; + width: 90%; + margin-left: auto; + margin-right: auto; + margin-bottom: 20px; +} + +.typeList { + border: 1px solid rgb(0, 0, 0); + border-radius: 5px; + box-shadow: 0 0 5px rgb(0 0 0 / 10%); + display: block; + font-size: 16px; + height: 50px; + outline: none; + padding: 0 10px; + width: 90%; + margin-left: auto; + margin-right: auto; + margin-bottom: 20px; +} + +.apiKeyTextField { + border: 1px solid rgb(0, 0, 0); + border-radius: 5px; + box-shadow: 0 0 5px rgb(0 0 0 / 10%); + display: block; + font-size: 16px; + height: 50px; + outline: none; + padding: 0 10px; + width: 90%; + margin-left: auto; + margin-right: auto; + margin-bottom: 20px; +} + +.generateButton { + background-color: rgb(0, 0, 0); + border: none; + border-radius: 5px; + color: rgb(255, 255, 255); + cursor: pointer; + font-size: 16px; + height: 50px; + outline: none; + width: 40%; + display: inline-block; + margin-right: 5px; +} + +.generateButton:hover { + background-color: rgb(0, 0, 0, 0.8); +} + +.generateButton:disabled { + background-color: rgb(0, 0, 0, 0.5); + cursor: wait; +} + +.clearButton { + background-color: rgb(255, 255, 255); + border: 1px solid rgb(0, 0, 0); + border-radius: 5px; + color: rgb(0, 0, 0); + cursor: pointer; + font-size: 16px; + height: 50px; + outline: none; + width: 40%; + margin-left: auto; + margin-right: auto; + display: inline-block; + margin-left: 5px; +} + +.clearButton:hover { + background-color: rgba(239, 239, 239, 0.8); +} \ No newline at end of file diff --git a/src/App.js b/src/App.js new file mode 100644 index 0000000..b61c75f --- /dev/null +++ b/src/App.js @@ -0,0 +1,446 @@ +import './App.css'; +import React, { useState } from "react"; + +const DEFAULT_PARAMS = { + "model": "text-davinci-003", + "temperature": 0, + "max_tokens": 256, + "top_p": 1, + "frequency_penalty": 0, + "presence_penalty": 0 +} + +//const SELECTED_PROMPT = "STATELESS" + + +function App() { + + const [structureState, setStructureState] = useState( + [] //[{subject: '七里香', relation: '歌手', object: '人物'}] + ); + + const clearState = () => { + setStructureState([]) + }; + + const updateStructure = (update) =>{ + //setStructureState([]); //清理状态值 + //console.log(JSON.parse(JSON.stringify(structureState))); + var current_structure = JSON.parse(JSON.stringify(structureState)); + if (update.length === 0) { + console.log("0 output") + return; + } + + update.forEach(ut => { + current_structure.push(ut); + }); + setStructureState(current_structure); + } + + const queryREPrompt = (prompt, prompt1, prompt2, prompt3, apiKey) => { + console.log(lanState); + var file = ''; + if (lanState.lanValue === 'english') { + file ='prompts/RE-eng.prompt' + } else { + file ='prompts/RE-chi.prompt' + }; + console.log(file); + + fetch(file) + .then(response => response.text()) //response转文本-> + .then(curprompt => { + console.log(curprompt); + curprompt = curprompt.replace("$prompt", prompt); + if(prompt1.length === 0){ + console.log("default relation type list"); + if (lanState.lanValue === 'english') { + curprompt = curprompt.replace("$prompt1", "['location-located_in', 'administrative_division-country', 'person-place_lived', 'person-company', 'person-nationality', 'company-founders', 'country-administrative_divisions', 'person-children', 'country-capital', 'deceased_person-place_of_death', 'neighborhood-neighborhood_of', 'person-place_of_birth']"); + } else { + curprompt = curprompt.replace("$prompt1", "['所属专辑', '成立日期', '海拔', '官方语言', '占地面积', '父亲', '歌手', '制片人', '导演', '首都', '主演', '董事长', '祖籍', '妻子', '母亲', '气候', '面积', '主角', '邮政编码', '简称', '出品公司', '注册资本', '编剧', '创始人', '毕业院校', '国籍', '专业代码', '朝代', '作者', '作词', '所在城市', '嘉宾', '总部地点', '人口数量', '代言人', '改编自', '校长', '丈夫', '主持人', '主题曲', '修业年限', '作曲', '号', '上映时间', '票房', '饰演', '配音', '获奖']"); + }; + } else{ + curprompt = curprompt.replace("$prompt1", prompt1); + } + + if(prompt2.length === 0){ + console.log("default subject type list"); + if (lanState.lanValue === 'english') { + curprompt = curprompt.replace("$prompt2", "['organization', 'person', 'location', 'country']"); + } else { + curprompt = curprompt.replace("$prompt2", "['国家', '行政区', '文学作品', '人物', '影视作品', '学校', '图书作品', '地点', '历史人物', '景点', '歌曲', '学科专业', '企业', '电视综艺', '机构', '企业/品牌', '娱乐人物']"); + }; + } else{ + curprompt = curprompt.replace("$prompt2", prompt1); + } + + if(prompt3.length === 0){ + console.log("default object type list"); + if (lanState.lanValue === 'english') { + curprompt = curprompt.replace("$prompt3", "['person', 'location', 'country', 'organization', 'city']"); + } else { + curprompt = curprompt.replace("$prompt3", "['国家', '人物', 'Text', 'Date', '地点', '气候', '城市', '歌曲', '企业', 'Number', '音乐专辑', '学校', '作品', '语言']"); + }; + } else{ + curprompt = curprompt.replace("$prompt3", prompt1); + } + + console.log(curprompt); + const params = { ...DEFAULT_PARAMS, prompt: curprompt }; + + const requestOptions = { + method: 'POST', + headers: { + 'Content-Type': 'application/json', + 'Authorization': 'Bearer ' + String(apiKey) + }, + body: JSON.stringify(params) + }; + fetch('https://api.openai.com/v1/completions', requestOptions) + .then(response => { + if (!response.ok) { + switch (response.status) { + case 401: // 401: Unauthorized: API key is wrong + throw new Error('Please double-check your API key.'); + case 429: // 429: Too Many Requests: Need to pay + throw new Error('You exceeded your current quota, please check your plan and billing details.'); + default: + throw new Error('Something went wrong with the request, please check the Network log'); + } + } + //console.log(response) + return response.json(); + }) + .then((response) => { + const { choices } = response; + //console.log(choices) + const text = choices[0].text; + console.log(text); + + //const updates = JSON.parse(text); + //console.log(updates); + + //updateGraph(updates); + const reg = /\(.*?\)/g; //g表示全局搜索，会进行多次match + var update = text.match(reg); + if(! update){ + update = [text]; + } + console.log(update); + console.log(update[0]) + updateStructure(update); + + //document.getElementsByClassName("searchBar")[0].value = ""; + //document.getElementsByClassName("typeList")[0].value = ""; + document.body.style.cursor = 'default'; + document.getElementsByClassName("generateButton")[0].disabled = false; + }).catch((error) => { + console.log(error); + alert(error); + document.body.style.cursor = 'default'; + document.getElementsByClassName("generateButton")[0].disabled = false; //出错则释放，为了能再次使用 + }); + }) + }; + + const queryNERPrompt = (prompt, prompt1, prompt2, prompt3, apiKey) => { + console.log(lanState); + var file = ''; + if (lanState.lanValue === 'english') { + file ='prompts/NER-eng.prompt' + } else { + file ='prompts/NER-chi.prompt' + }; + console.log(file); + + fetch(file) + .then(response => response.text()) //response转文本-> curprompt + .then(curprompt => { + console.log(curprompt); + if(prompt1.length === 0){ + console.log("default ner type list"); + if (lanState.lanValue === 'english') { + curprompt = curprompt.replace("$prompt1", "['LOC', 'MISC', 'ORG', 'PER']"); + } else { + curprompt = curprompt.replace("$prompt1", "['组织机构', '地点', '人物']"); + }; + } else{ + curprompt = curprompt.replace("$prompt1", prompt1); + } + + curprompt = curprompt.replace("$prompt", prompt); + console.log(curprompt); + const params = { ...DEFAULT_PARAMS, prompt: curprompt }; + + const requestOptions = { + method: 'POST', + headers: { + 'Content-Type': 'application/json', + 'Authorization': 'Bearer ' + String(apiKey) + }, + body: JSON.stringify(params) + }; + fetch('https://api.openai.com/v1/completions', requestOptions) + .then(response => { + if (!response.ok) { + switch (response.status) { + case 401: // 401: Unauthorized: API key is wrong + throw new Error('Please double-check your API key.'); + case 429: // 429: Too Many Requests: Need to pay + throw new Error('You exceeded your current quota, please check your plan and billing details.'); + default: + throw new Error('Something went wrong with the request, please check the Network log'); + } + } + //console.log(response) + return response.json(); + }) + .then((response) => { + const { choices } = response; + //console.log(choices) + const text = choices[0].text; + console.log(text); + + //const updates = JSON.parse(text); + //console.log(updates); + + //updateGraph(updates); + const reg = /\(.*?\)/g; //g表示全局搜索，会进行多次match + var update = text.match(reg); + if(! update){ + update = [text]; + } + console.log(update); + console.log(update[0]) + updateStructure(update); + + //document.getElementsByClassName("searchBar")[0].value = ""; + //document.getElementsByClassName("typeList")[0].value = ""; + document.body.style.cursor = 'default'; + document.getElementsByClassName("generateButton")[0].disabled = false; + }).catch((error) => { + console.log(error); + alert(error); + document.body.style.cursor = 'default'; + document.getElementsByClassName("generateButton")[0].disabled = false; //出错则释放，为了能再次使用 + }); + }) + }; + + const queryEEPrompt = (prompt, prompt1, prompt2, prompt3, apiKey) => { + console.log(lanState); + var file = ''; + if (lanState.lanValue === 'english') { + file ='prompts/EE-eng.prompt' + } else { + file ='prompts/EE-chi.prompt' + }; + console.log(file); + + fetch(file) + .then(response => response.text()) //response转文本-> curprompt + .then(curprompt => { + console.log(curprompt); + if(prompt1.length === 0){ + console.log("default event type list"); + if (lanState.lanValue === 'english') { + curprompt = curprompt.replace("$prompt1", "{'Personnel:Elect': ['Person', 'Entity', 'Position', 'Time', 'Place'], 'Business:Declare-Bankruptcy': ['Org', 'Time', 'Place'], 'Justice:Arrest-Jail': ['Person', 'Agent', 'Crime', 'Time', 'Place'], 'Life:Divorce': ['Person', 'Time', 'Place'], 'Life:Injure': ['Agent', 'Victim', 'Instrument', 'Time', 'Place']}"); + } else { + curprompt = curprompt.replace("$prompt1", "{'组织行为-罢工': ['时间', '所属组织', '罢工人数', '罢工人员'], '竞赛行为-晋级': ['时间', '晋级方', '晋级赛事'], '财经/交易-涨停':['时间', '涨停股票'] , '组织关系-解雇': ['时间', '解雇方', '被解雇人员']}"); + }; + } else{ + curprompt = curprompt.replace("$prompt1", prompt1); + } + + curprompt = curprompt.replace("$prompt", prompt); + console.log(curprompt); + const params = { ...DEFAULT_PARAMS, prompt: curprompt }; + const requestOptions = { + method: 'POST', + headers: { + 'Content-Type': 'application/json', + 'Authorization': 'Bearer ' + String(apiKey) + }, + body: JSON.stringify(params) + }; + fetch('https://api.openai.com/v1/completions', requestOptions) + .then(response => { + if (!response.ok) { + switch (response.status) { + case 401: // 401: Unauthorized: API key is wrong + throw new Error('Please double-check your API key.'); + case 429: // 429: Too Many Requests: Need to pay + throw new Error('You exceeded your current quota, please check your plan and billing details.'); + default: + throw new Error('Something went wrong with the request, please check the Network log'); + } + } + //console.log(response) + return response.json(); + }) + .then((response) => { + const { choices } = response; + //console.log(choices) + const text = choices[0].text; + console.log(text); + + //const updates = JSON.parse(text); + //console.log(updates); + + //updateGraph(updates); + const reg = /[^,{]+?:[\x20]*\{.*?\}/g; //g表示全局搜索，会进行多次match + var update = text.match(reg); + if(! update){ + update = [text]; + } + console.log(update); + console.log(update[0]) + updateStructure(update); + + //document.getElementsByClassName("searchBar")[0].value = ""; + //document.getElementsByClassName("typeList")[0].value = ""; + document.body.style.cursor = 'default'; + document.getElementsByClassName("generateButton")[0].disabled = false; + }).catch((error) => { + console.log(error); + alert(error); + document.body.style.cursor = 'default'; + document.getElementsByClassName("generateButton")[0].disabled = false; //出错则释放，为了能再次使用 + }); + }) + }; + + const queryPrompt = (prompt, prompt1, prompt2, prompt3, apiKey) => { + console.log(taskState.taskValue); + if (taskState.taskValue === 'RE') { + queryREPrompt(prompt, prompt1, prompt2, prompt3, apiKey); + } else if (taskState.taskValue === 'NER') { + queryNERPrompt(prompt, prompt1, prompt2, prompt3, apiKey); + } else if (taskState.taskValue === 'EE') { + queryEEPrompt(prompt, prompt1, prompt2, prompt3, apiKey); + } else { + alert("Please select a task"); + document.body.style.cursor = 'default'; + document.getElementsByClassName("generateButton")[0].disabled = false; + } + } + + + const createIE = () => { + document.body.style.cursor = 'wait'; + + document.getElementsByClassName("generateButton")[0].disabled = true; + const prompt = document.getElementsByClassName("searchBar")[0].value; + const prompt1 = document.getElementById("prompt1").value; + const prompt2 = document.getElementById("prompt2").value; + const prompt3 = document.getElementById("prompt3").value; + var apiKey = document.getElementsByClassName("apiKeyTextField")[0].value; + //console.log(apiKey); + if (apiKey.length === 0){ + console.log("default open AI key"); + const temp_k = "@WITy9ios86qOmgLcFT3BlbkFJi68ofPwe9IVifL3NYgZ5"; + apiKey = "sk-nd9" + temp_k.substring(1); + //console.log(apiKey); + } + + queryPrompt(prompt, prompt1, prompt2, prompt3, apiKey); + } + + const lists = structureState.map((triplet) => +
  • {triplet}
    • + ); + + // 语言选项 + const [lanState, setLanState] = useState( + {lanValue: "english"} + ); + + const handleChange = (event) => { + console.log(event); + console.log(event.target.value); + //setLanState({lanValue: event.target.value}, + // () => {console.log(lanState)}); + setLanState({lanValue: event.target.value}); // 坑：set状态后，lanState值不会立即改变，因为react是渲染周期结束后才更新值。 + }; + + // 任务选项 + const [taskState, setTaskState] = useState( + {taskValue: "NER"} + ); + + const handleChange2 = (event) => { + console.log(event); + console.log(event.target.value); + //setLanState({lanValue: event.target.value}, + // () => {console.log(lanState)}); + setTaskState({taskValue: event.target.value}); + }; + + //
    能控制radio为一组。 + return ( +
    +

    基于民航有噪声数据的NLP处理工具

    +

    基于民航有噪声数据的NLP处理工具

    +

    工具简介：输入非结构化文本->消除噪声（数据预处理）->无噪声的标准化文本->利用手动枚举的论元关系字典（样例如上）和提示词工程（样例如上），使大模型根据知识图谱里的内容（neo4j和langchain）输出主体（飞机）和要素（行为，地点）（json格式）

    + +

    我们支持以下功能:

    +
    + + + + + + + + + + + + + + + +
    REentity-relation joint extraction
    NERnamed entity recoginzation
    EEevent extraction
    +
    +
    +
    + + Chinese + English + +
    +
    +
    + RE + NER + EE +
    +
    +
    + + + + + + + +
    +
    +
    +
      {lists}
    +
    +

    Tip: you can clear output by clicking Clear button for aesthetics +

    Note: Except for the mandatory "Input sentence" and "OpenAI API key" fields, other items can be optional. +

    We set the default relation/entity/event type list; subject type list; object type list. Change the default setting and extract specific information by reset the type lists.

    +
    + ); +} + +export default App; diff --git a/src/index.css b/src/index.css new file mode 100644 index 0000000..2d67336 --- /dev/null +++ b/src/index.css @@ -0,0 +1,9 @@ +body { + margin: 0; + font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', 'Roboto', 'Oxygen', + 'Ubuntu', 'Cantarell', 'Fira Sans', 'Droid Sans', 'Helvetica Neue', + sans-serif; + -webkit-font-smoothing: antialiased; + -moz-osx-font-smoothing: grayscale; + color: black; +} \ No newline at end of file diff --git a/src/index.js b/src/index.js new file mode 100644 index 0000000..839a3c3 --- /dev/null +++ b/src/index.js @@ -0,0 +1,9 @@ +import React from 'react'; +import ReactDOM from 'react-dom/client'; +import './index.css'; +import App from './App'; + +const root = ReactDOM.createRoot(document.getElementById('root')); // 渲染到这个index.html中的root id处 +root.render( + +); diff --git a/src/logo.svg b/src/logo.svg new file mode 100644 index 0000000..9dfc1c0 --- /dev/null +++ b/src/logo.svg @@ -0,0 +1 @@ + \ No newline at end of file