Neighbors All Neighbors

Source header: cuvs/neighbors/all_neighbors.h

All-neighbors C-API build parameters

cuvsAllNeighborsAlgo

Graph build algorithm selection.

1 typedef enum {
2   CUVS_ALL_NEIGHBORS_ALGO_BRUTE_FORCE = 0,
3   CUVS_ALL_NEIGHBORS_ALGO_IVF_PQ = 1,
4   CUVS_ALL_NEIGHBORS_ALGO_NN_DESCENT = 2
5 } cuvsAllNeighborsAlgo;

Values

Name	Value
`CUVS_ALL_NEIGHBORS_ALGO_BRUTE_FORCE`	`0`
`CUVS_ALL_NEIGHBORS_ALGO_IVF_PQ`	`1`
`CUVS_ALL_NEIGHBORS_ALGO_NN_DESCENT`	`2`

cuvsAllNeighborsIndexParams

Parameters controlling SNMG all-neighbors build.

1 struct cuvsAllNeighborsIndexParams {
2   cuvsAllNeighborsAlgo algo;
3   size_t overlap_factor;
4   size_t n_clusters;
5   cuvsDistanceType metric;
6   cuvsIvfPqIndexParams_t ivf_pq_params;
7   cuvsNNDescentIndexParams_t nn_descent_params;
8 };

Fields

Name	Type	Description
`algo`	`cuvsAllNeighborsAlgo`
`overlap_factor`	`size_t`
`n_clusters`	`size_t`
`metric`	`cuvsDistanceType`
`ivf_pq_params`	`cuvsIvfPqIndexParams_t`
`nn_descent_params`	`cuvsNNDescentIndexParams_t`

cuvsAllNeighborsIndexParamsCreate

Create a default all-neighbors index parameters struct.

1 cuvsError_t cuvsAllNeighborsIndexParamsCreate(cuvsAllNeighborsIndexParams_t* index_params);

Parameters

Name	Direction	Type	Description
`index_params`	out	`cuvsAllNeighborsIndexParams_t*`	Pointer to allocated index_params struct

Returns

cuvsError_t

cuvsAllNeighborsIndexParamsDestroy

Destroy an all-neighbors index parameters struct.

1 cuvsError_t cuvsAllNeighborsIndexParamsDestroy(cuvsAllNeighborsIndexParams_t index_params);

Parameters

Name	Direction	Type	Description
`index_params`	in	`cuvsAllNeighborsIndexParams_t`	Index parameters struct to destroy

Returns

cuvsError_t

All-neighbors C-API build

cuvsAllNeighborsBuild

Build an all-neighbors k-NN graph automatically detecting host vs device dataset.

1 cuvsError_t cuvsAllNeighborsBuild(cuvsResources_t res,
2 cuvsAllNeighborsIndexParams_t params,
3 DLManagedTensor* dataset,
4 DLManagedTensor* indices,
5 DLManagedTensor* distances,
6 DLManagedTensor* core_distances,
7 float alpha);

The function automatically detects whether the dataset is host-resident or device-resident and calls the appropriate implementation. For host datasets, it partitions data into n_clusters clusters and assigns each row to overlap_factor nearest clusters. For device datasets, n_clusters must be 1 (no batching); overlap_factor is ignored. Outputs always reside in device memory.

Parameters

Name	Direction	Type	Description
`res`	in	`cuvsResources_t`	Can be a SNMG multi-GPU resources (`cuvsResources_t`) or single-GPU resources
`params`	in	`cuvsAllNeighborsIndexParams_t`	Build parameters (see cuvsAllNeighborsIndexParams)
`dataset`	in	`DLManagedTensor*`	2D tensor [num_rows x dim] on host or device (auto-detected)
`indices`	out	`DLManagedTensor*`	2D tensor [num_rows x k] on device (int64)
`distances`	out	`DLManagedTensor*`	Optional 2D tensor [num_rows x k] on device (float32); can be NULL
`core_distances`	out	`DLManagedTensor*`	Optional 1D tensor [num_rows] on device (float32); can be NULL
`alpha`	in	`float`	Mutual-reachability scaling; used only when core_distances is provided

Returns

cuvsError_t