Opensearch register and simulate opensearch-neural-sparse-encoding-multilingual-v1

opensearch.http

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"
  }
}


POST /_plugins/_ml/models/_register?deploy=true
{
  "name": "amazon/neural-sparse/opensearch-neural-sparse-encoding-multilingual-v1",
  "version": "1.0.0",
  "model_format": "TORCH_SCRIPT"
}

GET /_plugins/_ml/tasks/1fvAzpYBf6BSf_XtxoXh

#{
#  "model_id": "1vvAzpYBf6BSf_Xt0IU-",
#  "task_type": "REGISTER_MODEL",
# "function_name": "SPARSE_ENCODING",
# "state": "COMPLETED",
#  "worker_node": [
#    "g_XZWqxGSz2dQOgt1iDOkQ"
#  ],
# "create_time": 1747225462143,
#  "last_update_time": 1747225561551,
#  "is_async": true
#}


POST /_plugins/_ml/models/_search
{
}

PUT /_ingest/pipeline/sparse
{
  "description": "An sparse encoding ingest pipeline",
  "processors": [
    {
      "sparse_encoding": {
        "model_id": "1vvAzpYBf6BSf_Xt0IU-",
        "field_map": {
          "text": "embedding"
        }
      }
    }
  ]
}

POST /_ingest/pipeline/sparse/_simulate
{
  "docs": [
    {
      "_index": "index",
      "_id": "id",
      "_source": {
        "text": "document-name"
      }
    }
  ]
}