Shows how to use semantic search locally on OpenSearch using Distillibert.

local_semantic_search.md

Local Semantic Search in OpenSearch!

If you'd like to watch this demo through it's available on YouTube or if you prefer reading there is a walkthrough on my blog.

Cluster settings for Amazon OpenSearch and Locally running

PUT /_cluster/settings
{
    "persistent": {
        "plugins.ml_commons.only_run_on_ml_node": false,
        "plugins.ml_commons.model_access_control_enabled": true,
        "plugins.ml_commons.native_memory_threshold": "99",
        "plugins.ml_commons.model_auto_redeploy.enable": true,
        "plugins.ml_commons.model_auto_redeploy.lifetime_retry_times": 3
    }
}

Register your model group

POST /_plugins/_ml/model_groups/_register
{
    "name": "Model_Group",
    "description": "Public ML Model Group",
    "access_mode": "public"
}
#MODEL_GROUP: 

Register your model

POST /_plugins/_ml/models/_register
{
    "name": "huggingface/sentence-transformers/msmarco-distilbert-base-tas-b",
    "version": "1.0.2",
    "model_group_id": "<MODEL_GROUP>",
    "model_format": "TORCH_SCRIPT"
}
# TASK_ID:

Check if model is downloaded

GET /_plugins/_ml/tasks/<TASK_ID>
# MODEL_ID:

Deploy your model

POST /_plugins/_ml/models/<MODEL_ID>/_deploy

Check that model was deployed

GET /_plugins/_ml/tasks/<TASK_ID>

Testing the model

POST /_plugins/_ml/_predict/text_embedding/<MODEL_ID>
{
  "text_docs":[ "This should get embedded"],
  "return_number": true,
  "target_response": ["sentence_embedding"]
}

Create and ingestion pipeline for embeddings

PUT _ingest/pipeline/embedding-ingest-pipeline
{
  "description": "Neural Search Pipeline",
  "processors" : [
    {
      "text_embedding": {
        "model_id": "<MODEL_ID>",
        "field_map": {
          "content": "content_embedding"
        }
      }
    }
  ]
}

Creating a hybrid search pipeline

## Put the search pipeline in place
PUT _search/pipeline/hybrid-search-pipeline
{
  "phase_results_processors": [
    {
      "normalization-processor": {
        "normalization": {
          "technique": "min_max"
        },
        "combination": {
          "technique": "arithmetic_mean",
          "parameters": {
            "weights": [
              0.3,
              0.7
            ]
          }
        }
      }
    }
  ]
}

Create the index

PUT /documents
{
    "settings": {
        "index.knn": true,
        "default_pipeline": "embedding-ingest-pipeline",
        "index.search.default_pipeline": "hybrid-search-pipeline"
    },
    "mappings": {
        "properties": {
            "content_embedding": {
                "type": "knn_vector",
                "dimension": 768,
                "method": {
                    "name": "hnsw",
                    "space_type": "innerproduct",
                    "engine": "nmslib"
                }
            },
            "content": {
                "type": "text"
            }
        }
    }
}

Upload documents

POST /documents/_bulk
{ "index": {"_id": "1234" } }
{ "content": "There once was a racecar driver that was super fast"}
{ "index": {"_id": "1235" } }
{ "content": "The golf driver used by tiger woods is the TaylorMade Qi10 LS prototype"}
{ "index": {"_id": "1236" } }
{ "content": "Some may say that supercar drivers dont really mind risk"}

Search the documents with lexical and vector search

GET /documents/_search
{
  "_source": {
    "exclude": [
      "content_embedding"
    ]
  },
  "query": {
    "hybrid": {
      "queries": [
        {
          "match": {
            "content": {
              "query": "sports automobile"
            }
          }
        },
        {
          "neural": {
            "content_embedding": {
              "query_text": "sports cars",
              "model_id": "<MODEL_ID>",
              "k": 5
            }
          }
        }
      ]
    }
  }