Class CompressingQueryTransformer
-
- All Implemented Interfaces:
-
dev.langchain4j.rag.query.transformer.QueryTransformer
public class CompressingQueryTransformer implements QueryTransformer
A QueryTransformer that leverages a ChatLanguageModel to condense a given Query along with a chat memory (previous conversation history) into a concise Query. This is applicable only when a ChatMemory is in use. Refer to DEFAULT_PROMPT_TEMPLATE and implementation for more details. Configurable parameters (optional): - promptTemplate: The prompt template used to instruct the LLM to compress the specified Query.
-
-
Nested Class Summary
Nested Classes Modifier and Type Class Description public classCompressingQueryTransformer.CompressingQueryTransformerBuilder
-
Field Summary
Fields Modifier and Type Field Description public final static PromptTemplateDEFAULT_PROMPT_TEMPLATE
-
Constructor Summary
Constructors Constructor Description CompressingQueryTransformer(ChatLanguageModel chatLanguageModel)CompressingQueryTransformer(ChatLanguageModel chatLanguageModel, PromptTemplate promptTemplate)
-
Method Summary
Modifier and Type Method Description static CompressingQueryTransformer.CompressingQueryTransformerBuilderbuilder()Collection<Query>transform(Query query)Transforms the given Query into one or multiple Querys. -
-
Constructor Detail
-
CompressingQueryTransformer
CompressingQueryTransformer(ChatLanguageModel chatLanguageModel)
-
CompressingQueryTransformer
CompressingQueryTransformer(ChatLanguageModel chatLanguageModel, PromptTemplate promptTemplate)
-
-
Method Detail
-
builder
static CompressingQueryTransformer.CompressingQueryTransformerBuilder builder()
-
-
-
-