VectorIndex interface
Properties
| indexing |
This is the size of the candidate list of approximate neighbors stored while building the DiskANN index as part of the optimization processes. Large values may improve recall at the expense of latency. This is only applicable for the diskANN vector index type. |
| path | The path to the vector field in the document. |
| quantization |
The number of bytes used in product quantization of the vectors. A larger value may result in better recall for vector searches at the expense of latency. This is only applicable for the quantizedFlat and diskANN vector index types. |
| type | The index type of the vector. Currently, flat, diskANN, and quantizedFlat are supported. |
| vector |
Array of shard keys for the vector index. This is only applicable for the quantizedFlat and diskANN vector index types. |
Property Details
indexingSearchListSize
This is the size of the candidate list of approximate neighbors stored while building the DiskANN index as part of the optimization processes. Large values may improve recall at the expense of latency. This is only applicable for the diskANN vector index type.
indexingSearchListSize?: numberProperty Value
number
path
The path to the vector field in the document.
path: stringProperty Value
string
quantizationByteSize
The number of bytes used in product quantization of the vectors. A larger value may result in better recall for vector searches at the expense of latency. This is only applicable for the quantizedFlat and diskANN vector index types.
quantizationByteSize?: numberProperty Value
number
type
The index type of the vector. Currently, flat, diskANN, and quantizedFlat are supported.
type: stringProperty Value
string
vectorIndexShardKey
Array of shard keys for the vector index. This is only applicable for the quantizedFlat and diskANN vector index types.
vectorIndexShardKey?: string[]Property Value
string[]