Bring Your Own Embeddings - Weaviate Schema

This is our standard schema with bring your own vectors, where the Retrieval Suite generates the embeddings prior to indexing.

{
  "class": "__DOCUMENT_INDEX_NAME__",
  "name": "__DOCUMENT_INDEX_NAME__",
  "vectorIndexConfig": {
    "propertyName": "selfProvided"
  },
  "properties": [
    {
      "dataType": [
        "string[]"
      ],
      "name": "allow_acl",
      "indexFilterable": true,
      "tokenization": "field"
    },
    {
      "dataType": [
        "string[]"
      ],
      "name": "deny_acl",
      "indexFilterable": true,
      "tokenization": "field"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "authors",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "authorsMail",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "body",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "body_de",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "body_fr",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "body_es",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "body_it",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "fileExtension",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "lowercase"
    },
    {
      "dataType": [
        "date"
      ],
      "name": "createdDate",
      "indexFilterable": true
    },
    {
      "dataType": [
        "string"
      ],
      "name": "iconSvg",
      "indexFilterable": false,
      "tokenization": "field"
    },
    {
      "dataType": [
        "string"
      ],
      "name": "iconUrl",
      "indexFilterable": false,
      "tokenization": "field"
    },
     {
      "dataType": [
        "text"
      ],
      "name": "internalId",
      "indexFilterable": false,
      "indexSearchable": true,
      "tokenization": "field"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "itemType",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "lowercase"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "keywords",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "documentTopic",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "documentScope",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "documentCategory",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "persons",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "organizations",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "account",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "date"
      ],
      "name": "lastModifiedDate",
      "indexFilterable": true
    },
    {
      "dataType": [
        "text[]"
      ],
      "name": "originalItemType",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "lowercase"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "originalId",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "field"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "parentItemTitle",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "parentItemUrl",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "field"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "rootItemTitle",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "title",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "rootItemUrl",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "field"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "title_de",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "title_fr",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "title_es",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "title_it",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "word"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "url",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "field"
    },
    {
      "dataType": [
        "text"
      ],
      "name": "embedding",
      "indexFilterable": true,
      "indexSearchable": true,
      "tokenization": "field"
    }
  ]
}