opensearch-docs-cn/_search-plugins/neural-sparse-search.md

7.6 KiB

layout title nav_order has_children redirect_from
default Neural sparse search 50 false
/search-plugins/sparse-search/

Neural sparse search

Introduced 2.11 {: .label .label-purple }

Semantic search relies on dense retrieval that is based on text embedding models. However, dense methods use k-NN search, which consumes a large amount of memory and CPU resources. An alternative to semantic search, neural sparse search is implemented using an inverted index and is thus as efficient as BM25. Neural sparse search is facilitated by sparse embedding models. When you perform a neural sparse search, it creates a sparse vector (a list of token: weight key-value pairs representing an entry and its weight) and ingests data into a rank features index.

When selecting a model, choose one of the following options:

  • Use a sparse encoding model at both ingestion time and search time (high performance, relatively high latency).
  • Use a sparse encoding model at ingestion time and a tokenizer model at search time (low performance, relatively low latency).

PREREQUISITE
Before using neural sparse search, make sure to set up a pretrained sparse embedding model or your own sparse embedding model. For more information, see Choosing a model. {: .note}

To use neural sparse search, follow these steps:

  1. Create an ingest pipeline.
  2. Create an index for ingestion.
  3. Ingest documents into the index.
  4. Search the index using neural search.

Step 1: Create an ingest pipeline

To generate vector embeddings, you need to create an ingest pipeline that contains a sparse_encoding processor, which will convert the text in a document field to vector embeddings. The processor's field_map determines the input fields from which to generate vector embeddings and the output fields in which to store the embeddings.

The following example request creates an ingest pipeline where the text from passage_text will be converted into text embeddings and the embeddings will be stored in passage_embedding:

PUT /_ingest/pipeline/nlp-ingest-pipeline-sparse
{
  "description": "An sparse encoding ingest pipeline",
  "processors": [
    {
      "sparse_encoding": {
        "model_id": "aP2Q8ooBpBj3wT4HVS8a",
        "field_map": {
          "passage_text": "passage_embedding"
        }
      }
    }
  ]
}

{% include copy-curl.html %}

Step 2: Create an index for ingestion

In order to use the text embedding processor defined in your pipeline, create a rank features index, adding the pipeline created in the previous step as the default pipeline. Ensure that the fields defined in the field_map are mapped as correct types. Continuing with the example, the passage_embedding field must be mapped as rank_features. Similarly, the passage_text field should be mapped as text.

The following example request creates a rank features index that is set up with a default ingest pipeline:

PUT /my-nlp-index
{
  "settings": {
    "default_pipeline": "nlp-ingest-pipeline-sparse"
  },
  "mappings": {
    "properties": {
      "id": {
        "type": "text"
      },
      "passage_embedding": {
        "type": "rank_features"
      },
      "passage_text": {
        "type": "text"
      }
    }
  }
}

{% include copy-curl.html %}

To save disk space, you can exclude the embedding vector from the source as follows:

PUT /my-nlp-index
{
  "settings": {
    "default_pipeline": "nlp-ingest-pipeline-sparse"
  },
  "mappings": {
      "_source": {
      "excludes": [
        "passage_embedding"
      ]
    },
    "properties": {
      "id": {
        "type": "text"
      },
      "passage_embedding": {
        "type": "rank_features"
      },
      "passage_text": {
        "type": "text"
      }
    }
  }
}

{% include copy-curl.html %}

Once the <token, weight> pairs are excluded from the source, they cannot be recovered. Before applying this optimization, make sure you don't need the <token, weight> pairs for your application. {: .important}

Step 3: Ingest documents into the index

To ingest documents into the index created in the previous step, send the following requests:

PUT /my-nlp-index/_doc/1
{
  "passage_text": "Hello world",
  "id": "s1"
}

{% include copy-curl.html %}

PUT /my-nlp-index/_doc/2
{
  "passage_text": "Hi planet",
  "id": "s2"
}

{% include copy-curl.html %}

Before the document is ingested into the index, the ingest pipeline runs the sparse_encoding processor on the document, generating vector embeddings for the passage_text field. The indexed document includes the passage_text field, which contains the original text, and the passage_embedding field, which contains the vector embeddings.

To perform a neural sparse search on your index, use the neural_sparse query clause in Query DSL queries.

The following example request uses a neural_sparse query to search for relevant documents:

GET my-nlp-index/_search
{
  "query": {
    "neural_sparse": {
      "passage_embedding": {
        "query_text": "Hi world",
        "model_id": "aP2Q8ooBpBj3wT4HVS8a"
      }
    }
  }
}

{% include copy-curl.html %}

The response contains the matching documents:

{
  "took" : 688,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 2,
      "relation" : "eq"
    },
    "max_score" : 30.0029,
    "hits" : [
      {
        "_index" : "my-nlp-index",
        "_id" : "1",
        "_score" : 30.0029,
        "_source" : {
          "passage_text" : "Hello world",
          "passage_embedding" : {
            "!" : 0.8708904,
            "door" : 0.8587369,
            "hi" : 2.3929274,
            "worlds" : 2.7839446,
            "yes" : 0.75845814,
            "##world" : 2.5432441,
            "born" : 0.2682308,
            "nothing" : 0.8625516,
            "goodbye" : 0.17146169,
            "greeting" : 0.96817183,
            "birth" : 1.2788506,
            "come" : 0.1623208,
            "global" : 0.4371151,
            "it" : 0.42951578,
            "life" : 1.5750692,
            "thanks" : 0.26481047,
            "world" : 4.7300377,
            "tiny" : 0.5462298,
            "earth" : 2.6555297,
            "universe" : 2.0308156,
            "worldwide" : 1.3903781,
            "hello" : 6.696973,
            "so" : 0.20279501,
            "?" : 0.67785245
          },
          "id" : "s1"
        }
      },
      {
        "_index" : "my-nlp-index",
        "_id" : "2",
        "_score" : 16.480486,
        "_source" : {
          "passage_text" : "Hi planet",
          "passage_embedding" : {
            "hi" : 4.338913,
            "planets" : 2.7755864,
            "planet" : 5.0969057,
            "mars" : 1.7405145,
            "earth" : 2.6087382,
            "hello" : 3.3210192
          },
          "id" : "s2"
        }
      }
    ]
  }
}