PLAID

PLAID index with choice between fast-plaid (Rust-based) and Stanford NLP backends.

This class provides a unified interface for PLAID indexing that can use either: - FastPlaid: High-performance Rust-based implementation (default) - Stanford PLAID: Original Stanford NLP implementation (deprecated)

Parameters

• index_folder ('str') – defaults to indexes

  The folder where the index will be stored.

• index_name ('str') – defaults to colbert

  The name of the index.

• override ('bool') – defaults to False

  Whether to override the collection if it already exists.

• use_fast ('bool') – defaults to True

  If True (default), use fast-plaid backend. If False, use Stanford PLAID backend.

• nbits ('int') – defaults to 4

  The number of bits to use for product quantization. Lower values mean more compression and potentially faster searches but can reduce accuracy.

• kmeans_niters ('int') – defaults to 4

  The number of iterations for the K-means algorithm used during index creation. This influences the quality of the initial centroid assignments.

• max_points_per_centroid ('int') – defaults to 256

  The maximum number of points (token embeddings) that can be assigned to a single centroid during K-means. This helps in balancing the clusters.

• n_ivf_probe ('int') – defaults to 8

  The number of inverted file list "probes" to perform during the search. This parameter controls the number of clusters to search within the index for each query. Higher values improve recall but increase search time.

• n_full_scores ('int') – defaults to 8192

  The number of candidate documents for which full (re-ranked) scores are computed. This is a crucial parameter for accuracy; higher values lead to more accurate results but increase computation.

• n_samples_kmeans ('int | None') – defaults to None

  The number of samples to use for K-means clustering. If None, it defaults to a value based on the number of documents. This parameter can be adjusted to balance between speed, memory usage and clustering quality.

• batch_size ('int') – defaults to 262144

  The internal batch size used for processing queries. A larger batch size might improve throughput on powerful GPUs but can consume more memory.

• show_progress ('bool') – defaults to True

  If set to True, a progress bar will be displayed during search operations.

• device ('str | list[str] | None') – defaults to None

  Specifies the device(s) to use for computation. If None (default) and CUDA is available, it defaults to "cuda". If CUDA is not available, it defaults to "cpu". Can be a single device string (e.g., "cuda:0" or "cpu"). Can be a list of device strings (e.g., ["cuda:0", "cuda:1"]).

• use_triton ('bool | None') – defaults to None

  Whether to use triton kernels when computing kmeans using fast-plaid. Triton kernels are faster, but yields some variance due to race condition, set to false to get 100% reproducible results. If unset, will use triton kernels if possible.

• kwargs

Examples

>>> from pylate import indexes, models

>>> index = indexes.PLAID(
...    index_folder="test_index",
...    index_name="plaid_colbert",
...    override=True,
... )

>>> model = models.ColBERT(
...    model_name_or_path="lightonai/GTE-ModernColBERT-v1",
... )

>>> documents_embeddings = model.encode([
...    "Document content here...",
...    "Another document...",
... ] * 10, is_query=False)

>>> index = index.add_documents(
...    documents_ids=range(len(documents_embeddings)),
...    documents_embeddings=documents_embeddings
... )
Computing centroids of embeddings.
Creating FastPlaid index.

>>> queries_embeddings = model.encode(
...     ["search query", "hello world"],
...     is_query=True,
... )

>>> scores = index(
...     queries_embeddings,
...     k=10,
... )

>>> index = index.add_documents(
...    documents_ids=range(len(documents_embeddings), len(documents_embeddings) * 2),
...    documents_embeddings=documents_embeddings
... )

>>> scores = index(
...     queries_embeddings,
...     k=25,
... )

Methods

call

Query the index for the nearest neighbors of the query embeddings.

Parameters

• queries_embeddings ('np.ndarray | torch.Tensor')
• k ('int') – defaults to 10
• subset ('list[list[str]] | list[str] | None') – defaults to None

Returns

list[list[RerankResult]]: List of lists containing dictionaries with 'id' and 'score' keys.

add_documents

Add documents to the index.

Parameters

• documents_ids ('str | list[str]')
• documents_embeddings ('list[np.ndarray | torch.Tensor]')
• kwargs

get_documents_embeddings

Get document embeddings by their IDs.

Parameters

• document_ids ('list[list[str]]')

remove_documents

Remove documents from the index.

Parameters

• documents_ids ('list[str]')