{"id":9,"date":"2023-07-05T19:15:06","date_gmt":"2023-07-05T10:15:06","guid":{"rendered":"https:\/\/www.pinkgold.space\/?p=9"},"modified":"2024-09-23T17:16:19","modified_gmt":"2024-09-23T08:16:19","slug":"openai-platform-examples-qa","status":"publish","type":"post","link":"https:\/\/www.pinkgold.space\/?p=9","title":{"rendered":"OpenAI API Examples-Q&A"},"content":{"rendered":"\n

ChatGPT\u306eAPI\u3092\u697d\u3057\u3080\u70ba\u306b\u8ab2\u91d1\u3092\u3055\u308c\u305f\u65b9\u306f\u591a\u3044\u3093\u3058\u3083\u306a\u3044\u304b\u3068\u601d\u3044\u307e\u3059\u3002\u307e\u305f\u306f\u5371\u6a5f\u611f\u304c\u304d\u3063\u304b\u3051\u3068\u306a\u308a\u5b66\u3073\u306e\u305f\u3081\u306b\u8ab2\u91d1\u3055\u308c\u308b\u65b9\u3082\u3044\u308b\u304b\u3082\u3057\u308c\u307e\u305b\u3093\u3002\u3069\u306e\u307f\u3061\u8ab2\u91d1\u3092\u3057\u305f\u3089\u4f7f\u308f\u306a\u3044\u3068\u3082\u3063\u305f\u3044\u306a\u3044\u306e\u3067\u4f8b\u6587\u304f\u3089\u3044\u306f\u8a66\u3057\u3066\u697d\u3057\u3093\u3067\u304a\u304d\u305f\u3044\u3082\u306e\u3067\u3059\u3002<\/p>\n\n\n\n

\u307e\u305a\u306fExamples\u306e\u6700\u521d\u306b\u3042\u308bQ&A\u306b\u3064\u3044\u3066\u898b\u3066\u3044\u304d\u307e\u3057\u3087\u3046<\/p>\n\n\n\n

https:\/\/platform.openai.com\/examples\/default-qa<\/a> \u3067node.js\u3092\u9078\u629e\u3059\u308b\u3068\u30b5\u30f3\u30d7\u30eb\u306e\u30b3\u30fc\u30c9\u304c\u8868\u793a\u3055\u308c\u307e\u3059\u3002<\/p>\n\n\n\n

const { Configuration, OpenAIApi } = require(\"openai\");\n\nconst configuration = new Configuration({\n  apiKey: process.env.OPENAI_API_KEY,\n});\nconst openai = new OpenAIApi(configuration);\n\nconst response = await openai.createCompletion({\n  model: \"text-davinci-003\",\n  prompt: \"I am a highly intelligent question answering bot. If you ask me a question that is rooted in truth, I will give you the answer. If you ask me a question that is nonsense, trickery, or has no clear answer, I will respond with \\\"Unknown\\\".\\n\\nQ: What is human life expectancy in the United States?\\nA: Human life expectancy in the United States is 78 years.\\n\\nQ: Who was president of the United States in 1955?\\nA: Dwight D. Eisenhower was president of the United States in 1955.\\n\\nQ: Which party did he belong to?\\nA: He belonged to the Republican Party.\\n\\nQ: What is the square root of banana?\\nA: Unknown\\n\\nQ: How does a telescope work?\\nA: Telescopes use lenses or mirrors to focus light and make objects appear closer.\\n\\nQ: Where were the 1992 Olympics held?\\nA: The 1992 Olympics were held in Barcelona, Spain.\\n\\nQ: How many squigs are in a bonk?\\nA: Unknown\\n\\nQ: Where is the Valley of Kings?\\nA:\",\n  temperature: 0,\n  max_tokens: 100,\n  top_p: 1,\n  frequency_penalty: 0.0,\n  presence_penalty: 0.0,\n  stop: [\"\\n\"],\n});<\/code><\/pre>\n\n\n\n
\u3053\u306e\u307e\u307e\u30b3\u30d4\u30da\u3057\u3066\u3082\u52d5\u304d\u307e\u305b\u3093\u3002<\/h2>\n\n\n\n
OpenAI\u306e\u30d7\u30e9\u30c3\u30c8\u30d5\u30a9\u30fc\u30e0\u306e\u4f8b\u6587\u306f\u4e00\u822c\u7684\u306a\u4f7f\u7528\u4f8b\u3092\u793a\u3057\u3066\u3044\u308b\u3060\u3051\u3067\u3042\u308a\u3001\u5b8c\u5168\u306a\u52d5\u4f5c\u3092\u4fdd\u8a3c\u3059\u308b\u3082\u306e\u3067\u306f\u3042\u308a\u307e\u305b\u3093\u3002<\/p>\n\n\n\n
\u4f8b\u6587\u3092\u305d\u306e\u307e\u307e\u4f7f\u7528\u3059\u308b\u5834\u5408\u3001\u52d5\u4f5c\u3057\u306a\u3044\u53ef\u80fd\u6027\u304c\u3042\u308a\u307e\u3059\u3002\u305d\u308c\u306f\u3001\u4f8b\u6587\u304c\u7279\u5b9a\u306e\u8981\u4ef6\u3084\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u306b\u4f9d\u5b58\u3057\u3066\u3044\u308b\u304b\u3001\u30d0\u30fc\u30b8\u30e7\u30f3\u306e\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3084\u5909\u66f4\u306b\u8ffd\u968f\u3057\u3066\u3044\u306a\u3044\u53ef\u80fd\u6027\u304c\u3042\u308b\u304b\u3089\u3067\u3059\u3002<\/p>\n\n\n\n
\u6b63\u5e38\u306b\u52d5\u4f5c\u3055\u305b\u308b\u305f\u3081\u306b\u306f\u3001\u9069\u5207\u306aAPI\u547c\u3073\u51fa\u3057\u3068\u30c7\u30fc\u30bf\u51e6\u7406\u3092\u884c\u3046\u5fc5\u8981\u304c\u3042\u308a\u307e\u3059\u3002\u3053\u308c\u306b\u306f\u3001API\u306e\u30a8\u30f3\u30c9\u30dd\u30a4\u30f3\u30c8\u3001\u30ea\u30af\u30a8\u30b9\u30c8\u306e\u5f62\u5f0f\u3001\u5fc5\u8981\u306a\u30d1\u30e9\u30e1\u30fc\u30bf\u30fc\u3001\u304a\u3088\u3073\u30ec\u30b9\u30dd\u30f3\u30b9\u306e\u89e3\u91c8\u304c\u542b\u307e\u308c\u307e\u3059\u3002<\/p>\n\n\n\n
\u52d5\u4f5c\u74b0\u5883\u3092\u4f5c\u308a\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n
\u4e8b\u524d\u6e96\u5099\u3068\u3057\u3066node.js\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u307e\u3057\u3087\u3046\u3002\u516c\u5f0f\u30b5\u30a4\u30c8<\/a>\u304b\u3089LTS\u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3057\u3066\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3092\u3057\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n
\u4eca\u56de\u306fv18.16.0\u3092\u4f7f\u3044\u307e\u3059\u3002<\/p>\n\n\n\n
package.json\u306f\u6b21\u306e\u901a\u308a\u3067\u3059\u3002\u4f5c\u6210\u3057\u305f\u3089npm install\u3067\u5fc5\u8981\u306a\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u307e\u3059\u3002<\/p>\n\n\n\n
[user@somewhere stub]$ cat package.json\n{\n  \"dependencies\": {\n    \"dotenv\": \"^16.3.1\",\n    \"openai\": \"^3.3.0\"\n  }\n}\n[user@somewhere stub]$ npm install<\/code><\/pre>\n\n\n\n
\u52d5\u304f\u3088\u3046\u306b\u4fee\u6b63\u3057\u3066\u307f\u307e\u3057\u3087\u3046\u3002<\/h2>\n\n\n\n
\u4e8b\u524d\u6e96\u5099\u3068\u3057\u3066.env\u30d5\u30a1\u30a4\u30eb\u3092\u7528\u610f\u3057\u767b\u9332\u3057\u305fAPI\u306e\u30c8\u30fc\u30af\u30f3\u3092\u8a2d\u5b9a\u3057\u3066\u304a\u304d\u307e\u3059\u3002<\/p>\n\n\n\n
[user@somewhrere stub]$ cat .env\nOPENAI_API_KEY=xxxxxxxxxxxxxxxxxxxxxxxxxxx<\/code><\/pre>\n\n\n\n
\u65e9\u901f\u5b9f\u88c5\u3057\u3066\u3044\u304d\u307e\u3059\u3002API\u30ad\u30fc\u3092\u8aad\u307f\u8fbc\u3093\u3067\u51e6\u7406\u306e\u90e8\u5206\u3092\u95a2\u6570\u306b\u3057\u3066\u3044\u304d\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n
require('dotenv').config();  \/\/ dotenv\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30dd\u30fc\u30c8\nconst { Configuration, OpenAIApi } = require(\"openai\");\n\nconst configuration = new Configuration({\n  apiKey: process.env.OPENAI_API_KEY,\n});\nconst openai = new OpenAIApi(configuration);\n\nasync function run(){\n    const response = await openai.createCompletion({\n        model: \"text-davinci-003\",\n       prompt: \"I am a highly intelligent question answering bot. If you ask me a question that is rooted in truth, I will give you the answer. If you ask me a question that is nonsense, trickery, or has no clear answer, I will respond with \\\"Unknown\\\".\\n\\nQ: What is human life expectancy in the United States?\\nA: Human life expectancy in the United States is 78 years.\\n\\nQ: Who was president of the United States in 1955?\\nA: Dwight D. Eisenhower was president of the United States in 1955.\\n\\nQ: Which party did he belong to?\\nA: He belonged to the Republican Party.\\n\\nQ: What is the square root of banana?\\nA: Unknown\\n\\nQ: How does a telescope work?\\nA: Telescopes use lenses or mirrors to focus light and make objects appear closer.\\n\\nQ: Where were the 1992 Olympics held?\\nA: The 1992 Olympics were held in Barcelona, Spain.\\n\\nQ: How many squigs are in a bonk?\\nA: Unknown\\n\\nQ: Where is the Valley of Kings?\\nA:\",\n        temperature: 0,\n        max_tokens: 100,\n        top_p: 1,\n        frequency_penalty: 0.0,\n        presence_penalty: 0.0,\n        stop: [\"\\n\"],\n    });\n    if (response.data && response.data.choices && response.data.choices.length > 0) {\n      console.log(response.data.choices[0].text);\n    } else {\n      console.log(\"No response received or empty choices.\");\n    }\n}\n<\/code><\/pre>\n\n\n\n
\u5b9f\u884c\u3059\u308b\u3068\u6b21\u306e\u3088\u3046\u306b\u7d50\u679c\u3092\u5f97\u3089\u308c\u307e\u3059\u3002<\/p>\n\n\n\n
\n[user@somewhere stub]$ node qa.js\n The Valley of Kings is located in Luxor, Egypt.<\/code><\/pre>\n\n\n\n
\u305d\u308c\u3067\u306f\u6df1\u5800\u3057\u3066\u3044\u304d\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n
createCompletion\u95a2\u6570\u306b\u6e21\u3057\u3066\u3044\u308b\u5f15\u6570\u304b\u3089\u898b\u3066\u3044\u304d\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n
\u307e\u305a\u306fmodel: “text-davinci-003”,\u306e\u90e8\u5206\u3067\u3059\u3002chatGPT\u306b\u306f\u8907\u6570\u306e\u30e2\u30c7\u30eb\u304c\u5b58\u5728\u3057\u307e\u3059\u3002<\/p>\n\n\n\n
\u516c\u5f0f\u30b5\u30a4\u30c8\u306eOverView\u3092\u898b\u3066\u307f\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n
URL\u306fhttps:\/\/platform.openai.com\/docs\/models\/overview<\/a>\u3067\u3059\u3002<\/a><\/p>\n\n\n\n
Overview<\/a><\/h2>\n\n\n\n
MODELS<\/th>DESCRIPTION<\/th><\/tr><\/thead>
GPT-4<\/a><\/td>A set of models that improve on GPT-3.5 and can understand as well as generate natural language or code
GPT-3.5\u3092\u6539\u826f\u3057\u3001\u81ea\u7136\u8a00\u8a9e\u3084\u30b3\u30fc\u30c9\u3092\u7406\u89e3\u30fb\u751f\u6210\u3067\u304d\u308b\u30e2\u30c7\u30eb\u306e\u96c6\u5408\u3002<\/td><\/tr>
GPT-3.5<\/a><\/td>A set of models that improve on GPT-3 and can understand as well as generate natural language or code
GPT-3 \u3092\u6539\u826f\u3057\u3001\u81ea\u7136\u8a00\u8a9e\u3084\u30b3\u30fc\u30c9\u3092\u7406\u89e3\u30fb\u751f\u6210\u3067\u304d\u308b\u30e2\u30c7\u30eb\u7fa4<\/td><\/tr>
DALL\u00b7E<\/a><\/td>A model that can generate and edit images given a natural language prompt
\u81ea\u7136\u8a00\u8a9e\u306e\u30d7\u30ed\u30f3\u30d7\u30c8\u3092\u4e0e\u3048\u3066\u753b\u50cf\u3092\u751f\u6210\u30fb\u7de8\u96c6\u3067\u304d\u308b\u30e2\u30c7\u30eb<\/td><\/tr>
Whisper<\/a><\/td>A model that can convert audio into text
\u97f3\u58f0\u3092\u30c6\u30ad\u30b9\u30c8\u306b\u5909\u63db\u3067\u304d\u308b\u30e2\u30c7\u30eb<\/td><\/tr>
Embeddings<\/a><\/td>A set of models that can convert text into a numerical form
\u30c6\u30ad\u30b9\u30c8\u3092\u6570\u5024\u306b\u5909\u63db\u3067\u304d\u308b\u30e2\u30c7\u30eb\u306e\u30bb\u30c3\u30c8<\/td><\/tr>
Moderation<\/a><\/td>A fine-tuned model that can detect whether text may be sensitive or unsafe
\u7e4a\u7d30\u306a\u30c6\u30ad\u30b9\u30c8\u304b\u5b89\u5168\u3067\u306a\u3044\u30c6\u30ad\u30b9\u30c8\u304b\u3092\u691c\u51fa\u3067\u304d\u308b\u3001\u5fae\u8abf\u6574\u3055\u308c\u305f\u30e2\u30c7\u30eb<\/td><\/tr>
GPT-3<\/a>Legacy<\/td>A set of models that can understand and generate natural language
\u81ea\u7136\u8a00\u8a9e\u3092\u7406\u89e3\u3057\u751f\u6210\u3067\u304d\u308b\u30e2\u30c7\u30eb\u306e\u30bb\u30c3\u30c8<\/td><\/tr>
Deprecated<\/a><\/td>A full list of models that have been deprecated
\u975e\u63a8\u5968\u3068\u306a\u3063\u305f\u30e2\u30c7\u30eb\u306e\u5168\u30ea\u30b9\u30c8<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n
\u4eca\u56de\u306e\u4f8b\u6587\u3067\u767b\u5834\u3059\u308b\u306e\u306ftext-davinci-003\u3067\u3057\u305f\u3002\u3069\u3046\u3044\u3046\u30e2\u30c7\u30eb\u306a\u306e\u304b\u898b\u3066\u3044\u304d\u307e\u3057\u3087\u3046\u3002 curie, babbage, or ada\u3088\u308a\u3082\u512a\u308c\u305f\u54c1\u8cea\u3068\u3042\u308a\u307e\u3059\u304c\u3001\u305d\u3082\u305d\u3082\u305d\u306e3\u4eba\u306e\u3053\u3068\u304c\u30d4\u30f3\u3068\u304d\u307e\u305b\u3093\u3002turbo\u304c\u3064\u3044\u3066\u3044\u308b\u30e2\u30c7\u30eb\u306f\u9032\u5316\u3092\u3064\u3065\u3051\u3066\u3044\u308b\u3088\u3046\u3067\u3059\u306d\u3002<\/p>\n\n\n\n
LATEST MODEL<\/th>DESCRIPTION<\/th>MAX TOKENS<\/th>TRAINING DATA<\/th><\/tr><\/thead>
gpt-3.5-turbo<\/td>Most capable GPT-3.5 model and optimized for chat at 1\/10th the cost of text-davinci-003<\/code>. Will be updated with our latest model iteration 2 weeks after it is released.
text-davinci-003\u306e1\/10\u306e\u30b3\u30b9\u30c8\u3067\u30c1\u30e3\u30c3\u30c8\u306b\u6700\u9069\u5316\u3055\u308c\u305fGPT-3.5\u6700\u9ad8\u6027\u80fd\u30e2\u30c7\u30eb\u3002\u6700\u65b0\u30e2\u30c7\u30eb\u306f\u767a\u58f2\u5f8c2\u9031\u9593\u3067\u66f4\u65b0\u3055\u308c\u307e\u3059\u3002<\/td>		4,096 tokens<\/td>Up to Sep 2021<\/td><\/tr>
gpt-3.5-turbo-16k<\/td>Same capabilities as the standard gpt-3.5-turbo<\/code> model but with 4 times the context.
\u6a19\u6e96\u306egpt-3.5\u30bf\u30fc\u30dc\u30fb\u30e2\u30c7\u30eb\u3068\u540c\u3058\u6027\u80fd\u3060\u304c\u3001\u30b3\u30f3\u30c6\u30af\u30b9\u30c8\u306f4\u500d\u3002<\/td>		16,384 tokens<\/td>Up to Sep 2021<\/td><\/tr>
gpt-3.5-turbo-0613<\/td>Snapshot of gpt-3.5-turbo<\/code> from June 13th 2023 with function calling data. Unlike gpt-3.5-turbo<\/code>, this model will not receive updates, and will be deprecated 3 months after a new version is released.
2023\u5e746\u670813\u65e5\u306egpt-3.5-turbo\u306e\u30b9\u30ca\u30c3\u30d7\u30b7\u30e7\u30c3\u30c8\u3002gpt-3.5-turbo\u3068\u306f\u7570\u306a\u308a\u3001\u3053\u306e\u30e2\u30c7\u30eb\u306f\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3092\u53d7\u3051\u305a\u3001\u65b0\u3057\u3044\u30d0\u30fc\u30b8\u30e7\u30f3\u304c\u30ea\u30ea\u30fc\u30b9\u3055\u308c\u3066\u304b\u30893\u30f6\u6708\u5f8c\u306b\u975e\u63a8\u5968\u3068\u306a\u308a\u307e\u3059\u3002<\/td>		4,096 tokens<\/td>Up to Sep 2021<\/td><\/tr>
gpt-3.5-turbo-16k-0613<\/td>Snapshot of gpt-3.5-turbo-16k<\/code> from June 13th 2023. Unlike gpt-3.5-turbo-16k<\/code>, this model will not receive updates, and will be deprecated 3 months after a new version is released.
2023\u5e746\u670813\u65e5\u306egpt-3.5-turbo-16k\u306e\u30b9\u30ca\u30c3\u30d7\u30b7\u30e7\u30c3\u30c8\u3002gpt-3.5-turbo-16k\u3068\u306f\u7570\u306a\u308a\u3001\u3053\u306e\u30e2\u30c7\u30eb\u306f\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3092\u53d7\u3051\u305a\u3001\u65b0\u3057\u3044\u30d0\u30fc\u30b8\u30e7\u30f3\u304c\u30ea\u30ea\u30fc\u30b9\u3055\u308c\u3066\u304b\u30893\u30f6\u6708\u5f8c\u306b\u975e\u63a8\u5968\u3068\u306a\u308a\u307e\u3059\u3002<\/td>		16,384 tokens<\/td>Up to Sep 2021<\/td><\/tr>
text-davinci-003 (Legacy)<\/td>Can do any language task with better quality, longer output, and consistent instruction-following than the curie, babbage, or ada models. Also supports some additional features such as inserting text<\/a>.
\u30ad\u30e5\u30ea\u30fc\u3001\u30d0\u30d9\u30c3\u30b8\u3001\u30a2\u30c0\u306e\u5404\u30e2\u30c7\u30eb\u3088\u308a\u3082\u512a\u308c\u305f\u54c1\u8cea\u3001\u9577\u3044\u51fa\u529b\u3001\u4e00\u8cab\u3057\u305f\u547d\u4ee4\u8ffd\u5f93\u6027\u3067\u3042\u3089\u3086\u308b\u8a00\u8a9e\u30bf\u30b9\u30af\u3092\u3053\u306a\u3059\u3053\u3068\u304c\u3067\u304d\u308b\u3002\u307e\u305f\u3001\u30c6\u30ad\u30b9\u30c8\u306e\u633f\u5165\u306a\u3069\u3001\u3044\u304f\u3064\u304b\u306e\u8ffd\u52a0\u6a5f\u80fd\u3082\u30b5\u30dd\u30fc\u30c8\u3057\u3066\u3044\u308b\u3002
<\/td>		4,097 tokens<\/td>Up to Jun 2021<\/td><\/tr>
text-davinci-002 (Legacy)<\/td>Similar capabilities to text-davinci-003<\/code> but trained with supervised fine-tuning instead of reinforcement learning
text-davinci-003\u3068\u540c\u69d8\u306e\u6a5f\u80fd\u3092\u6301\u3064\u304c\u3001\u5f37\u5316\u5b66\u7fd2\u306e\u4ee3\u308f\u308a\u306b\u6559\u5e2b\u3042\u308a\u306e\u5fae\u8abf\u6574\u3067\u5b66\u7fd2\u3055\u305b\u308b\u3002<\/td>		4,097 tokens<\/td>Up to Jun 2021<\/td><\/tr>
code-davinci-002 (Legacy)<\/td>Optimized for code-completion tasks
\u30b3\u30fc\u30c9\u88dc\u5b8c\u30bf\u30b9\u30af\u306e\u6700\u9069\u5316<\/td>		8,001 tokens<\/td>Up to Jun 2021<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n
\u6b21\u306b\u7b2c\u4e8c\u5f15\u6570\u306eprompt\u3067\u3059\u304c\u3001\u8cea\u554f\u306a\u3093\u3060\u306a\u3068\u308f\u304b\u308b\u306e\u3067\u98db\u3070\u3057\u307e\u3059\u3002<\/p>\n\n\n\n
\u7b2c\u4e09\u5f15\u6570\u306etemperature: 0,\u306e\u90e8\u5206\u3092\u898b\u3066\u3044\u304d\u307e\u3057\u3087\u3046\u3002temperature\u3068\u306f\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u306b\u304a\u3051\u308b\u6e29\u5ea6\u306e\u3088\u3046\u3067\u3059\u306d\u3002<\/p>\n\n\n\n
\u6e29\u5ea6\u306f0\u304b\u30892\u306e\u9593\u30020.8\u306e\u3088\u3046\u306a\u9ad8\u3044\u5024\u306f\u51fa\u529b\u3092\u3088\u308a\u30e9\u30f3\u30c0\u30e0\u306b\u3057\u30010.2\u306e\u3088\u3046\u306a\u4f4e\u3044\u5024\u306f\u51fa\u529b\u3092\u3088\u308a\u96c6\u4e2d\u3057\u3066\u6c7a\u5b9a\u8ad6\u7684\u306b\u3059\u308b\u3002\u3068\u3042\u308a\u307e\u3059\u3002\u3053\u308c\u3060\u3051\u3067\u3082\u3044\u308d\u3044\u308d\u697d\u3057\u3081\u305d\u3046\u3067\u3059\u3002<\/p>\n\n\n\n
\n
(property) CreateCompletionRequest[‘temperature’]?: number | null | undefined<\/p>\n\n\n\n
What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p<\/code> but not both.<\/p>\n\n\n\n
@type<\/em> \u2014 {number}<\/p>\n\n\n\n
@memberof<\/em> \u2014 CreateCompletionRequest<\/p>\n<\/blockquote>\n\n\n\n
<\/p>\n\n\n\n
\u7b2c\u56db\u5f15\u6570\u306emax_tokens\u306f\u76f4\u89b3\u7684\u306b\u7406\u89e3\u3067\u304d\u308b\u306e\u3067\u98db\u3070\u3057\u307e\u3059\u3002<\/p>\n\n\n\n
\u7b2c\u4e94\u5f15\u6570\u306e top_p: 1\u306b\u3064\u3044\u3066\u307f\u3066\u3044\u304d\u307e\u3057\u3087\u3046<\/p>\n\n\n\n
\u300c\u6838\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u3068\u547c\u3070\u308c\u308b\u3001\u6e29\u5ea6\u306b\u3088\u308b\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u306b\u4ee3\u308f\u308b\u3082\u306e\u3067\u3001top_p \u306e\u78ba\u7387\u8cea\u91cf\u3092\u6301\u3064\u30c8\u30fc\u30af\u30f3\u306e\u7d50\u679c\u3092\u8003\u616e\u3059\u308b\u30e2\u30c7\u30eb\u3067\u3059\u3002\u3064\u307e\u308a\u30010.1\u306f\u4e0a\u4f4d10%\u306e\u78ba\u7387\u3092\u69cb\u6210\u3059\u308b\u30c8\u30fc\u30af\u30f3\u306e\u307f\u304c\u8003\u616e\u3055\u308c\u308b\u3053\u3068\u3092\u610f\u5473\u3057\u307e\u3059\u3002\u4e00\u822c\u7684\u306b\u306f\u3001\u3053\u308c\u304b\u6e29\u5ea6\u3092\u5909\u66f4\u3059\u308b\u3053\u3068\u3092\u63a8\u5968\u3057\u307e\u3059\u304c\u3001\u4e21\u65b9\u306f\u63a8\u5968\u3057\u307e\u305b\u3093\u3002\u300d<\/p>\n\n\n\n
\n
(property) CreateCompletionRequest[‘top_p’]?: number | null | undefined
An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both.<\/p>\n\n\n\n
@type \u2014 {number}<\/p>\n\n\n\n
@memberof \u2014 CreateCompletionRequest<\/p>\n<\/blockquote>\n\n\n\n
\u3053\u3053\u3067\u78ba\u7387\u8cea\u91cf\u3068\u3044\u3046\u7528\u8a9e\u304c\u3067\u3066\u304d\u307e\u3059\u3002\u78ba\u7387\u8cea\u91cf\u95a2\u6570\u306b\u3064\u3044\u3066\u304b\u307f\u7815\u3044\u3066\u77e5\u308a\u305f\u3044\u65b9\u306f<\/p>\n\n\n\n
\u99c6\u3051\u51fa\u3057\u30a8\u30f3\u30b8\u30cb\u30a2\u3055\u3093\u306e\u52d5\u753b<\/a>\u304c\u53c2\u8003\u306b\u306a\u308b\u3068\u601d\u3044\u307e\u3059\u3002<\/p>\n\n\n\n
<\/p>\n\n\n\n
\u7b2c\u516d\u5f15\u6570\u306efrequency_penalty: 0.0\u306b\u3064\u3044\u3066\u307f\u3066\u3044\u304d\u307e\u3057\u3087\u3046<\/p>\n\n\n\n
\n
Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model’s likelihood to repeat the same line verbatim. See more information about frequency and presence penalties.<\/a><\/p>\n\n\n\n
@type<\/em> \u2014 {number}<\/p>\n\n\n\n
@memberof<\/em> \u2014 CreateCompletionRequest<\/p>\n<\/blockquote>\n\n\n\n
2.0\u304b\u30892.0\u306e\u9593\u306e\u6570\u5024\u3002\u6b63\u306e\u5024\u306f\u3001\u65b0\u3057\u3044\u30c8\u30fc\u30af\u30f3\u306b\u3001\u3053\u308c\u307e\u3067\u306e\u30c6\u30ad\u30b9\u30c8\u306b\u304a\u3051\u308b\u65e2\u5b58\u306e\u983b\u5ea6\u306b\u57fa\u3065\u3044\u3066\u30da\u30ca\u30eb\u30c6\u30a3\u3092\u4e0e\u3048\u3001\u30e2\u30c7\u30eb\u304c\u540c\u3058\u884c\u3092\u9010\u8a9e\u7684\u306b\u7e70\u308a\u8fd4\u3059\u53ef\u80fd\u6027\u3092\u6e1b\u3089\u3057\u307e\u3059\u3002<\/p>\n\n\n\n
\u7b2c\u4e03\u5f15\u6570\u306epresence_penalty: 0.0\u306b\u3064\u3044\u3066\u3082\u540c\u69d8\u3067\u3059\u306d\u3002<\/p>\n\n\n\n
\n
Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model’s likelihood to talk about new topics. See more information about frequency and presence penalties.<\/a><\/p>\n\n\n\n
@type<\/em> \u2014 {number}<\/p>\n\n\n\n
@memberof<\/em> \u2014 CreateCompletionRequest<\/p>\n<\/blockquote>\n\n\n\n
\u4e00\u756a\u6700\u5f8c\u306e\u7b2c8\u5f15\u6570 stop: [“\\n”]\u306e\u90e8\u5206\u3092\u898b\u3066\u3044\u304d\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n
\u3053\u308c\u306f\u5358\u7d14\u306b\u8cea\u554f\u306f\u3053\u3053\u307e\u3067\u3060\u3088\u3068\u3001\u533a\u5207\u308a\u6587\u5b57\u3092\u4f1d\u3048\u308b\u3082\u306e\u3067\u3059\u3002<\/p>\n\n\n\n
\u6700\u5f8c\u306b<\/p>\n\n\n\n
\u4f8b\u6587\u4ee5\u5916\u306b\u3082\u5f15\u6570\u304c\u3042\u308b\u306e\u3067\u516c\u5f0f\u30b5\u30a4\u30c8\u306eAPI\u30ea\u30d5\u30a1\u30ec\u30f3\u30b9\u3092\u3088\u304f\u8aad\u3093\u3067\u7406\u89e3\u3057\u307e\u3057\u3087\u3046\u3002<\/p>\n\n\n\n
https:\/\/platform.openai.com\/docs\/api-reference\/completions<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"
ChatGPT\u306eAPI\u3092\u697d\u3057\u3080\u70ba\u306b\u8ab2\u91d1\u3092\u3055\u308c\u305f\u65b9\u306f\u591a\u3044\u3093\u3058\u3083\u306a\u3044\u304b\u3068\u601d\u3044\u307e\u3059\u3002\u307e\u305f\u306f\u5371\u6a5f\u611f\u304c\u304d\u3063\u304b\u3051\u3068\u306a\u308a\u5b66\u3073\u306e\u305f\u3081\u306b\u8ab2\u91d1\u3055\u308c\u308b\u65b9\u3082\u3044\u308b\u304b\u3082\u3057\u308c\u307e\u305b\u3093\u3002\u3069\u306e\u307f\u3061\u8ab2\u91d1\u3092\u3057\u305f\u3089\u4f7f\u308f\u306a\u3044\u3068\u3082\u3063\u305f\u3044\u306a\u3044\u306e\u3067\u4f8b\u6587\u304f\u3089\u3044\u306f\u8a66\u3057\u3066\u697d\u3057 … \u7d9a\u304d\u3092\u8aad\u3080<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[],"class_list":["post-9","post","type-post","status-publish","format-standard","hentry","category-openapi"],"_links":{"self":[{"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=\/wp\/v2\/posts\/9"}],"collection":[{"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=9"}],"version-history":[{"count":8,"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=\/wp\/v2\/posts\/9\/revisions"}],"predecessor-version":[{"id":31,"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=\/wp\/v2\/posts\/9\/revisions\/31"}],"wp:attachment":[{"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=9"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=9"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.pinkgold.space\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=9"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}