Spring/spring-ai
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix TokenCountBatchingStrategy example. (#1702)

Browse Source
This commit is contained in:
heqiang
2024-11-20 02:10:21 +08:00
committed by GitHub
parent c529c839ee
commit d2e9e55d25
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
spring-ai-docs/src/main/antora/modules/ROOT/pages/api/vectordbs.adoc
View File

@@ -142,7 +142,7 @@ public class EmbeddingConfig {
return new TokenCountBatchingStrategy(
EncodingType.CL100K_BASE, // Specify the encoding type
8000, // Set the maximum input token count
0.9 // Set the threshold factor
0.1 // Set the reserve percentage
);
}
}
@@ -152,7 +152,7 @@ In this configuration:
1. `EncodingType.CL100K_BASE`: Specifies the encoding type used for tokenization. This encoding type is used by the `JTokkitTokenCountEstimator` to accurately estimate token counts.
2. `8000`: Sets the maximum input token count. This value should be less than or equal to the maximum context window size of your embedding model.
3. `0.9`: Sets the threshold factor. This factor determines how full a batch can be before starting a new one. A value of 0.9 means each batch will be filled up to 90% of the maximum input token count.
3. `0.1`: Sets the reserve percentage. The percentage of tokens to reserve from the max input token count. This creates a buffer for potential token count increases during processing.
By default, this constructor uses `Document.DEFAULT_CONTENT_FORMATTER` for content formatting and `MetadataMode.NONE` for metadata handling. If you need to customize these parameters, you can use the full constructor with additional parameters.