👐 Pre-filtering

Pre-filtering is a technique to optimize vector search by only considering documents that match certain criteria during vector search.

Fill in any <CODE_BLOCK_N> placeholders and run the cells under the Step 7: Adding pre-filters to your vector search section in the notebook to experiment with combining pre-filters with your vector search queries.

The answers for code blocks in this section are as follows:

CODE_BLOCK_12

Answer

{
    "name": ATLAS_VECTOR_SEARCH_INDEX_NAME,
    "type": "vectorSearch",
    "definition": {
        "fields": [
            {
                "type": "vector",
                "path": "embedding",
                "numDimensions": 512,
                "similarity": "cosine",
            },
            {"type": "filter", "path": "year"},
        ]
    },
}

CODE_BLOCK_13

Answer

{"year": {"$gte": 2002}}

CODE_BLOCK_14

Answer

{
    "name": ATLAS_VECTOR_SEARCH_INDEX_NAME,
    "type": "vectorSearch",
    "definition": {
        "fields": [
            {
                "type": "vector",
                "path": "embedding",
                "numDimensions": 512,
                "similarity": "cosine",
            },
            {"type": "filter", "path": "year"},
            {"type": "filter", "path": "pages"},
        ]
    },
}

CODE_BLOCK_15

Answer

{"$and": [{"year": {"$gte": 2002}}, {"pages": {"$lte": 250}}]}