본문 바로가기

ML&DL and LLM

LangChain

 

참조 : https://python.langchain.com/docs/modules/

 

Modules | 🦜️🔗 Langchain

LangChain provides standard, extendable interfaces and external integrations for the following main modules:

python.langchain.com

 

Modules

구분 Modules Components Item-1 Item-2 Link
Main 1.
Model I/O
1.1 Concepts 1.1
1.2
Prompts
1.2.1
Template

Composition
PromptTemplate
ChatPromptTemplate
1.2.1
String prompt composition
Chat prompt composition
1.2.2
Example Selector
Similarity
MMR
Length
Ngram
1.2.2
1.2.3
Custom Example Selector
 
1.2.4
Few-shot prompt templates
Few-shot examples for chat models
1.2.4
1.2.5
MessagePromptTemplate
AIMessagePromptTemplate
SystemMessagePromptTemplate
HumanMessagePromptTemplate
1.2.5
1.2.6
Partial prompt templates
Partial with strings
Partial with functions
 
1.2.7
Pipeline
Final prompt
Pipeline prompts
 
1.3
LLMs
1.3.1
Quick Start
Setup
LCEL
1.3.1
1.3.2
Custom LLM
   
1.3.3
Caching

Streaming

Tracking token usage
InMemoryCache
SQLiteCache
1.3.3
 
 
1.4
ChatModels
1.4.1
Quick Start
Setup
Messages
LCEL
1.4.1
Function calling Binding functions
Defining functions schemas
 
Caching In Memory Cache
SQLite Cache
1.3.3 Caching과
동일
Custom Chat Model    
Get log probabilities    
Streaming    
Tracking token usage    
1.5
Output Parsers
1.5.1
Types
CSV parser
Datetime parser
Enum parser
JSON parser
OpenAI Funtions
OpenAI Tools
Output-fixing parser
Pandas DataFrame Parser
Pydantic parser
Retry parser 
Strunctred output parser
XML parser
YAML parser
1.5.1
2.
Retrieval
2.1
Concept
2.1
2.2
Document loaders
Type CSV
File Directory
HTML
JSON
Markdown
PDF
2.2
Text Splitters Type HTMLHeaderTextSplitter
Split by character
Split code
MarkdownHeaderTextSplitter
Recursively split JSON
Recursively split by character
Semantic Chunking
Split by tokens
2.3
Text embedding
models
    2.4
Vector stores Get started Similarity search
Similarity search by vector
2.5
Asynchronous operations Similarity search
Similarity search by vector
 
MMR    
Retrievers Type Vector store-backed retriever
MultiQueryRetriever
Contextual comparession
Ensemble Retriever
Long-Context Reorder
MultiVector Retriever
Parent Document Retriever
Self-querying
Time-weighted vector store retriever
2.6
Indexing      
3.
Agents
Agent Types
Tools
     
Additional Chain        
Memory        
Callbacks        

 

'ML&DL and LLM' 카테고리의 다른 글

LangChain - 1.2.1 PromptTemplate  (0) 2024.03.27
LangChain - 1.1 Model I/O Concept  (0) 2024.03.27
[Langchain] Prompt 튜닝  (1) 2024.03.05
[ML&DL] AI - 2 - AI 역사  (1) 2019.04.20
[ML&DL] AI - 1 - AI란 무엇인가?  (0) 2019.04.20