Add docstring for encode_kwargs

Author: lsz05

src/jmteb/embedders/base.py

@@ -25,6 +25,7 @@ def encode(self, text: str | list[str], prefix: str | None = None, **kwargs) ->
         Args:
             text (str | list[str]): text string, or a list of texts.
             prefix (str, optional): the prefix to use for encoding. Default to None.
+            **kwargs: some more settings that may be necessary for specific models.
         """
         raise NotImplementedError
 

src/jmteb/evaluators/classification/evaluator.py

@@ -29,6 +29,7 @@ class ClassificationEvaluator(EmbeddingEvaluator):
         classifiers (dict[str, Classifier]): classifiers to be evaluated.
         prefix (str | None): prefix for sentences. Defaults to None.
         log_predictions (bool): whether to log predictions of each datapoint.
+        encode_kwargs (dict): kwargs passed to embedder's encode function. Defaults to {}.
     """
 
     def __init__(

src/jmteb/evaluators/clustering/evaluator.py

@@ -24,6 +24,14 @@
 class ClusteringEvaluator(EmbeddingEvaluator):
     """
     ClusteringEvaluator is a class for evaluating clustering models.
+
+    Args:
+        val_dataset (ClusteringDataset): validation dataset
+        test_dataset (ClusteringDataset): evaluation dataset
+        prefix (str | None): prefix for sentences. Defaults to None.
+        random_seed (int | None): random seed used in clustering models. Defaults to None.
+        log_predictions (bool): whether to log predictions of each datapoint.
+        encode_kwargs (dict): kwargs passed to embedder's encode function. Defaults to {}.
     """
 
     def __init__(

src/jmteb/evaluators/pair_classification/evaluator.py

@@ -22,6 +22,7 @@ class PairClassificationEvaluator(EmbeddingEvaluator):
         test_dataset (PairClassificationDataset): test dataset
         sentence1_prefix (str | None): prefix for sentence1. Defaults to None.
         sentence2_prefix (str | None): prefix for sentence2. Defaults to None.
+        encode_kwargs (dict): kwargs passed to embedder's encode function. Default to {}.
 
     # NOTE: Don't log predictions, as predictions by different metrics could be different.
     """

src/jmteb/evaluators/reranking/evaluator.py

@@ -39,6 +39,8 @@ class RerankingEvaluator(EmbeddingEvaluator):
         doc_prefix (str | None): prefix for documents. Defaults to None.
         log_predictions (bool): whether to log predictions of each datapoint. Defaults to False.
         top_n_docs_to_log (int): log only top n documents. Defaults to 5.
+        query_encode_kwargs (dict): kwargs passed to embedder's encode function when encoding queries. Defaults to {}.
+        doc_encode_kwargs (dict): kwargs passed to embedder's encode function when encoding documents. Defaults to {}.
     """
 
     def __init__(

src/jmteb/evaluators/retrieval/evaluator.py

@@ -42,6 +42,8 @@ class RetrievalEvaluator(EmbeddingEvaluator):
         doc_prefix (str | None): prefix for documents. Defaults to None.
         log_predictions (bool): whether to log predictions of each datapoint. Defaults to False.
         top_n_docs_to_log (int): log only top n documents that are predicted as relevant. Defaults to 5.
+        query_encode_kwargs (dict): kwargs passed to embedder's encode function when encoding queries. Defaults to {}.
+        doc_encode_kwargs (dict): kwargs passed to embedder's encode function when encoding documents. Defaults to {}.
     """
 
     def __init__(

src/jmteb/evaluators/sts/evaluator.py

@@ -26,6 +26,7 @@ class STSEvaluator(EmbeddingEvaluator):
         test_dataset (STSDataset): test dataset
         sentence1_prefix (str | None): prefix for sentence1. Defaults to None.
         sentence2_prefix (str | None): prefix for sentence2. Defaults to None.
+        encode_kwargs (dict): kwargs passed to embedder's encode function. Defaults to {}.
     """
 
     def __init__(