Variable reconfigureConst

Type declaration

• invertedIndex: ((options) => InvertedIndexConfigUpdate)

  Create an InvertedIndexConfigUpdate object to be used when updating the configuration of the keyword searching algorithm of your collection.

  See the docs for details!

  • • (options): InvertedIndexConfigUpdate

    • Parameters

      • options: {
            bm25b?: number;
            bm25k1?: number;
            cleanupIntervalSeconds?: number;
            stopwordsAdditions?: string[];
            stopwordsPreset?: "none" | "en";
            stopwordsRemovals?: string[];
        }

        • Optional bm25b?: number

          The BM25 b parameter.

        • Optional bm25k1?: number

          The BM25 k1 parameter.

        • Optional cleanupIntervalSeconds?: number

          The interval in seconds at which the inverted index is cleaned up.

        • Optional stopwordsAdditions?: string[]

          Additional stopwords to add.

        • Optional stopwordsPreset?: "none" | "en"

          The stopwords preset to use.

        • Optional stopwordsRemovals?: string[]

          Stopwords to remove.

      Returns InvertedIndexConfigUpdate

• replication: ((options) => RecursivePartial<ReplicationConfig>)

  Create a ReplicationConfigUpdate object to be used when defining the replication configuration of Weaviate.

  See the docs for more details.

  • • (options): RecursivePartial<ReplicationConfig>

    • Parameters

      • options: {
            asyncEnabled?: boolean;
            factor?: number;
        }

        • Optional asyncEnabled?: boolean

          Whether asynchronous replication is enabled.

        • Optional factor?: number

          The replication factor.

      Returns RecursivePartial<ReplicationConfig>

• vectorIndex: {
      flat: ((options) => ModuleConfig<"flat", VectorIndexConfigFlatUpdate>);
      hnsw: ((options) => ModuleConfig<"hnsw", VectorIndexConfigHNSWUpdate>);
      quantizer: {
          bq: ((options?) => BQConfigUpdate);
          pq: ((options?) => PQConfigUpdate);
          sq: ((options?) => SQConfigUpdate);
      };
  }

  • flat: ((options) => ModuleConfig<"flat", VectorIndexConfigFlatUpdate>)

    Create a ModuleConfig<'flat', VectorIndexConfigFlatUpdate> object to update the configuration of the FLAT vector index.

    Use this method when defining the options.vectorIndexConfig argument of the reconfigure.vectorizer method.

    Returns

    The configuration object.

    • • (options): ModuleConfig<"flat", VectorIndexConfigFlatUpdate>

      • Parameters

        • options: {
              quantizer?: BQConfigUpdate;
              vectorCacheMaxObjects?: number;
          }

          • Optional quantizer?: BQConfigUpdate

            The quantizer configuration to use. Default is bq.

          • Optional vectorCacheMaxObjects?: number

            The maximum number of objects to cache in the vector cache. Default is 1000000000000.

        Returns ModuleConfig<"flat", VectorIndexConfigFlatUpdate>

  • hnsw: ((options) => ModuleConfig<"hnsw", VectorIndexConfigHNSWUpdate>)

    Create a ModuleConfig<'hnsw', VectorIndexConfigHNSWCreate> object to update the configuration of the HNSW vector index.

    Use this method when defining the options.vectorIndexConfig argument of the reconfigure.vectorizer method.

    Returns

    The configuration object.

    • • (options): ModuleConfig<"hnsw", VectorIndexConfigHNSWUpdate>

      • Parameters

        • options: {
              dynamicEfFactor?: number;
              dynamicEfMax?: number;
              dynamicEfMin?: number;
              ef?: number;
              flatSearchCutoff?: number;
              quantizer?: PQConfigUpdate | BQConfigUpdate | SQConfigUpdate;
              vectorCacheMaxObjects?: number;
          }

          • Optional dynamicEfFactor?: number

            The dynamic ef factor. Default is 8.

          • Optional dynamicEfMax?: number

            The dynamic ef max. Default is 500.

          • Optional dynamicEfMin?: number

            The dynamic ef min. Default is 100.

          • Optional ef?: number

            The ef parameter. Default is -1.

          • Optional flatSearchCutoff?: number

            The flat search cutoff. Default is 40000.

          • Optional quantizer?: PQConfigUpdate | BQConfigUpdate | SQConfigUpdate

            The quantizer configuration to use. Use vectorIndex.quantizer.bq or vectorIndex.quantizer.pq to make one.

          • Optional vectorCacheMaxObjects?: number

            The maximum number of objects to cache in the vector cache. Default is 1000000000000.

        Returns ModuleConfig<"hnsw", VectorIndexConfigHNSWUpdate>

  • quantizer: {
        bq: ((options?) => BQConfigUpdate);
        pq: ((options?) => PQConfigUpdate);
        sq: ((options?) => SQConfigUpdate);
    }

    Define the quantizer configuration to use when creating a vector index.

    • bq: ((options?) => BQConfigUpdate)

      Create an object of type BQConfigUpdate to be used when updating the quantizer configuration of a vector index.

      NOTE: If the vector index already has a quantizer configured, you cannot change its quantizer type; only its values. So if you want to change the quantizer type, you must recreate the collection.

      Returns

      The configuration object.

      • • (options?): BQConfigUpdate

        • Parameters

          • Optional options: {
                cache?: boolean;
                rescoreLimit?: number;
            }

            • Optional cache?: boolean

              Whether to cache the quantizer.

            • Optional rescoreLimit?: number

              The new rescore limit.

          Returns BQConfigUpdate

    • pq: ((options?) => PQConfigUpdate)

      Create an object of type PQConfigUpdate to be used when updating the quantizer configuration of a vector index.

      NOTE: If the vector index already has a quantizer configured, you cannot change its quantizer type; only its values. So if you want to change the quantizer type, you must recreate the collection.

      Returns

      The configuration object.

      • • (options?): PQConfigUpdate

        • Parameters

          • Optional options: {
                centroids?: number;
                pqEncoderDistribution?: PQEncoderDistribution;
                pqEncoderType?: PQEncoderType;
                segments?: number;
                trainingLimit?: number;
            }

            • Optional centroids?: number

              The new number of centroids.

            • Optional pqEncoderDistribution?: PQEncoderDistribution

              The new encoder distribution.

            • Optional pqEncoderType?: PQEncoderType

              The new encoder type.

            • Optional segments?: number

              The new number of segments.

            • Optional trainingLimit?: number

              The new training limit.

          Returns PQConfigUpdate

    • sq: ((options?) => SQConfigUpdate)

      Create an object of type SQConfigUpdate to be used when updating the quantizer configuration of a vector index.

      NOTE: If the vector index already has a quantizer configured, you cannot change its quantizer type; only its values. So if you want to change the quantizer type, you must recreate the collection.

      Returns

      The configuration object.

      • • (options?): SQConfigUpdate

        • Parameters

          • Optional options: {
                rescoreLimit?: number;
                trainingLimit?: number;
            }

            • Optional rescoreLimit?: number

              The rescore limit.

            • Optional trainingLimit?: number

              The training limit.

          Returns SQConfigUpdate

• vectorizer: {
      update: (<N, I>(options) => VectorConfigUpdate<N, I>);
  }

  • update: (<N, I>(options) => VectorConfigUpdate<N, I>)

    Create a VectorConfigUpdate object to be used when updating the named vector configuration of Weaviate.

    • • <N, I>(options): VectorConfigUpdate<N, I>

      • Type Parameters

        • N extends undefined | string
        • I extends string

        Parameters

        • options: VectorizerUpdateOptions<N, I>

          The options for the named vector.

        Returns VectorConfigUpdate<N, I>