Train Index

This function is only present in the core library version of CyborgDB. In other versions (microservice, serverless), it is automatically called once enough vector embeddings have been indexed.

Builds the index using the specified training configuration. Required before efficient querying. Prior to calling this, all queries will be conducted using encrypted exhaustive search. After, they will be conducted using encrypted ANN search.

def train(self, batch_size: int = 2048, max_iters: int = 100, tolerance: float = 1e-6, max_memory: int = 0)

Parameters

Parameter	Type	Default	Description
`batch_size`	`int`	`2048`	(Optional) Size of each batch for training.
`max_iters`	`int`	`100`	(Optional) Maximum number of iterations for training.
`tolerance`	`float`	`1e-6`	(Optional) Convergence tolerance for training.
`max_memory`	`int`	`0`	(Optional) Maximum memory to use for training. `0` sets no limit.

There must be at least 2 * n_lists vector embeddings in the index prior to to calling this function.

Exceptions

ValueError

Raises an exception if the index was not created or loaded yet.

RuntimeError

Raises an exception if there are not enough vector embeddings in the index for training (must be at least 2 * n_lists).
Raises an exception if the index could not be trained.

Example Usage

# Load index
index = client.load_index(index_name=index_name, index_key=index_key)

# Train the index with custom settings
index.train(batch_size=128, max_iters=10, tolerance=1e-4)

# Train with default settings (auto-selected configuration)
index.train()

Introduction

Client

Encrypted Index

Types & Getters

Parameters

Exceptions

Example Usage

​Parameters

​Exceptions

​Example Usage

Parameters

Exceptions

Example Usage