{"id":63499,"date":"2023-05-13T16:58:54","date_gmt":"2023-05-13T07:58:54","guid":{"rendered":"https:\/\/smilegate.ai\/?p=63499"},"modified":"2023-05-13T16:58:56","modified_gmt":"2023-05-13T07:58:56","slug":"langchain-if-you-know-chatgpt-without-question","status":"publish","type":"post","link":"https:\/\/smilegate.ai\/cn\/2023\/05\/13\/langchain-if-you-know-chatgpt-without-question\/","title":{"rendered":"LangChain: If you know ChatGPT, without question"},"content":{"rendered":"
[\uac00\uc0c1\uc0dd\uba85\uc5f0\uad6c\ud300 \ud669\uc900\uc120]<\/p>\n\n\n\n
ChatGPT\uc640 Bard \ub4f1, \uc694\uc998 \ub300\ud654\ud615\ud0dc\uc758 \ub300\ud615 \uc5b8\uc5b4 \ubaa8\ub378(LLM)\uc774 \uc6b0\ud6c4\uc8fd\uc21c \ubc1c\ud45c\ub418\uace0 \uc788\ub2e4. \ud558\uc9c0\ub9cc, LLM\ub9cc \uc788\ub2e4\uba74 \ud559\uc2b5\ud55c \ub370\uc774\ud130 \uc548\uc5d0\uc11c\ub9cc \uc801\uc808\ud55c \ubb38\uc7a5\uc744 \uc0dd\uc131\ud574\ub0bc \uac83\uc774\ub2e4. \uadf8\ub798\uc11c Bard\ub294 \uad6c\uae00 \uac80\uc0c9 \uc5d4\uc9c4\uc744 \ucd94\uac00\ud558\uc5ec \ucd5c\uadfc \ub370\uc774\ud130\ub97c \ud65c\uc6a9\ud55c \ub2f5\ubcc0\uc744 \uc0dd\uc131\ud558\ub294 \uac15\uc810\uc744 \uac00\uc9c0\uace0 \uc788\ub2e4. \uc774\uac83\uc744 \uc6b0\ub9ac\ub3c4 \ud560 \uc218 \uc5c6\uc744\uae4c? \ub9cc\uc57d \ud604\uc7ac ChatGPT\uc640 Bard\ub97c \ubaa8\ub450 \uc368\ubcf4\uace0 \uc790\uc2e0\uc758 \uc751\uc6a9 \ud504\ub85c\uadf8\ub7a8\uc5d0 \uc801\uc6a9\ud558\uace0 \uc2f6\ub2e4\uba74, LangChain\uc774\ub77c\ub294 \uc624\ud508\uc18c\uc2a4\ub97c \ub208\uc5ec\uaca8 \ubcfc \ud544\uc694\uac00 \uc788\ub2e4. \uc774\ubc88 \ud3ec\uc2a4\ud2b8\ub294 LangChain\uc758 \uae30\ubcf8 \uc694\uc18c\ub4e4\uc744 \uc18c\uac1c\ud558\uace0, \uac04\ub2e8\ud55c \ucf54\ub4dc\ub97c \uc608\uc2dc\ub85c \uc2e4\ud589 \uacb0\uacfc\ub97c \ubcf4\uc5ec\uc904 \uac83\uc774\ub2e4.<\/p>\n\n\n\n
1. Moduls<\/strong><\/p>\n\n\n\n 1.1. Models<\/strong><\/p>\n\n\n\n 1.2. Prompts<\/strong><\/p>\n\n\n\n 1.3. Indexes<\/strong><\/p>\n\n\n\n 1.4. Memory<\/strong><\/p>\n\n\n\n 1.5. Chains<\/strong><\/p>\n\n\n\n 1.6. Agents<\/strong><\/p>\n\n\n\n 2. Example Code<\/p>\n\n\n\n 3. Execute Results<\/p>\n\n\n\n 4. Todo<\/p>\n\n\n\n <\/p>\n\n\n\n <\/p>\n [\uac00\uc0c1\uc0dd\uba85\uc5f0\uad6c\ud300 \ud669\uc900\uc120] ChatGPT\uc640 Bard \ub4f1, \uc694\uc998 \ub300\ud654\ud615\ud0dc\uc758 \ub300\ud615 \uc5b8\uc5b4 \ubaa8\ub378(LLM)\uc774 \uc6b0\ud6c4\uc8fd\uc21c \ubc1c\ud45c\ub418\uace0 \uc788\ub2e4. \ud558\uc9c0\ub9cc, LLM\ub9cc \uc788\ub2e4\uba74 \ud559\uc2b5\ud55c \ub370\uc774\ud130 \uc548\uc5d0\uc11c\ub9cc \uc801\uc808\ud55c \ubb38\uc7a5\uc744 \uc0dd\uc131\ud574\ub0bc \uac83\uc774\ub2e4. \uadf8\ub798\uc11c Bard\ub294 \uad6c\uae00 \uac80\uc0c9 \uc5d4\uc9c4\uc744 \ucd94\uac00\ud558\uc5ec \ucd5c\uadfc \ub370\uc774\ud130\ub97c \ud65c\uc6a9\ud55c \ub2f5\ubcc0\uc744 \uc0dd\uc131\ud558\ub294 \uac15\uc810\uc744 \uac00\uc9c0\uace0 \uc788\ub2e4. \uc774\uac83\uc744 \uc6b0\ub9ac\ub3c4 \ud560 \uc218 \uc5c6\uc744\uae4c? \ub9cc\uc57d \ud604\uc7ac ChatGPT\uc640 Bard\ub97c \ubaa8\ub450 \uc368\ubcf4\uace0 \uc790\uc2e0\uc758 \uc751\uc6a9 \ud504\ub85c\uadf8\ub7a8\uc5d0 \uc801\uc6a9\ud558\uace0 \uc2f6\ub2e4\uba74, LangChain\uc774\ub77c\ub294 \uc624\ud508\uc18c\uc2a4\ub97c…<\/p>\n\n
\n
\n
<\/figure>\n\n\n\n
\n
\n
\n
\n
<\/figure>\n\n\n\n
\n
\n
\n\n\n\n#!\/usr\/bin\/env python\r\n# -*- coding: utf-8 -*-\r\n\r\nimport os\r\n\r\nfrom langchain.agents import AgentType, initialize_agent\r\nfrom langchain.llms import OpenAI\r\nfrom langchain.memory import ConversationBufferWindowMemory\r\nfrom langchain.tools import Tool\r\nfrom langchain.utilities import GoogleSearchAPIWrapper\r\n\r\nos.environ[\r\"OPENAI_API_KEY\"] = \"\"\r\nos.environ[\"GOOGLE_API_KEY\"] = \"\"\r\nos.environ[\"GOOGLE_CSE_ID\"] = \"\"\r\n\r\nllm = OpenAI(temperature=0)\r\n\r\nsearch = GoogleSearchAPIWrapper()\r\ntools = [\r\n Tool(name=\"Google Search\",\r\n func=search.run,\r\n description=\"Search Google for recent results.\")\r\n]\r\n\r\nmemory = ConversationBufferWindowMemory(memory_key=\"chat_history\",\r\n k=5,\r\n return_messages=True)\r\n\r\nagent_chain = initialize_agent(\r\n tools,\r\n llm,\r\n agent=AgentType.CHAT_CONVERSATIONAL_REACT_DESCRIPTION,\r\n verbose=True,\r\n memory=memory)\r\n\r\r\nwhile True:\r\n agent_chain.run(input('Instruction: '))\r<\/code><\/pre>\n\n\n\n
\n\n\n\n<\/figure>\n\n\n\n
\n\n\n\n\n
\n\n\n\n