model

Union of string and path

Default: "cross-encoder/ms-marco-MiniLM-L-6-v2"

The name or path of a pre-trained cross-encoder model from Hugging Face.
Reqiured.

device

ComponentDevice

Default: None

The device on which the model is loaded. If None, the default device is automatically selected.
Optional.

token

Secret

Default: {"type": "env_var", "env_vars": ["HF_API_TOKEN"], "strict": False}

The API token used to download private models from Hugging Face.
Optional.

top_k

Int

Default: 10

The maximum number of Documents to return per query. Required.

query_prefix

String

Default: "" (empty string)

A string to add to the beginning of the query text before ranking. Can be used to prepend the text with an instruction, as required by some reranking models, such as bge.
Required.

document_prefix

String

Default: "" (empty string)

A string to add to the beginning of each document text before ranking. Can be used to prepend the text with an instruction, as required by some embedding models, such as bge.
Required.

meta_fields_to_embed

List of strings

Default: None

List of metadata fields that should be embedded along with the document content.
Optional.

embedding_separator

String

Default: "\n"

Separator used to concatenate the meta fields to the document content.
Required.

scale_score

Bool

True, False
Default: True

Whether the raw logit predictions should be scaled using a Sigmoid activation function. Set this to False if you don't want any scaling of the raw logit predictions.
Required.

calibration_factor

Float

Default: 1.0

The factor used for calibrating probabilities calculated by sigmoid(logits * calibration_factor). This is only used if scale_score is set to True.
Optional.

score_threshold

Float

Default: None

If provided, only returns documents with a score above this threshold.
Optional.

model_kwargs

Dictionary of string and any

Default: None

Additional keyword arguments passed to AutoModelForSequenceClassification.from_pretrained when loading the model specified in model.
Optional.