# Generated by SnakeBridge v0.14.0 - DO NOT EDIT MANUALLY
# Regenerate with: mix compile
# Library: dspy 3.1.2
# Python module: dspy.predict
# Python class: KNN

defmodule Dspy.Predict.KNNClass do
  @moduledoc """
  Wrapper for Python class KNN.
  """
  def __snakebridge_python_name__, do: "dspy.predict"
  def __snakebridge_python_class__, do: "KNN"
  def __snakebridge_library__, do: "dspy"
  @opaque t :: SnakeBridge.Ref.t()

  @doc """
  A k-nearest neighbors retriever that finds similar examples from a training set.



  ## Parameters

  - `k` - Number of nearest neighbors to retrieve
  - `trainset` - List of training examples to search through
  - `vectorizer` - The `Embedder` to use for vectorization

  ## Examples

      ```python
      import dspy
      from sentence_transformers import SentenceTransformer

      # Create a training dataset with examples
      trainset = [
      dspy.Example(input="hello", output="world"),
      # ... more examples ...
      ]

      # Initialize KNN with a sentence transformer model
      knn = KNN(
      k=3,
      trainset=trainset,
      vectorizer=dspy.Embedder(SentenceTransformer("all-MiniLM-L6-v2").encode)
      )

      # Find similar examples
      similar_examples = knn(input="hello")
      ```
  """
  @spec new(
          integer(),
          list(Dspy.Primitives.ExampleClass.t()),
          Dspy.Clients.Embedding.Embedder.t(),
          keyword()
        ) :: {:ok, SnakeBridge.Ref.t()} | {:error, Snakepit.Error.t()}
  def new(k, trainset, vectorizer, opts \\ []) do
    SnakeBridge.Runtime.call_class(__MODULE__, :__init__, [k, trainset, vectorizer], opts)
  end
end