handle default embedder config usage

Author: voorhs

src/autointent/_pipeline/_pipeline.py

@@ -19,6 +19,7 @@
     InferenceNodeConfig,
     LoggingConfig,
     VectorIndexConfig,
+    get_default_embedder_config,
     get_default_vector_index_config,
 )
 from autointent.custom_types import ListOfGenericLabels, NodeType, SearchSpacePreset, SearchSpaceValidationMode
@@ -56,7 +57,7 @@ def __init__(
 
         if isinstance(nodes[0], NodeOptimizer):
             self.logging_config = LoggingConfig()
-            self.embedder_config = EmbedderConfig()
+            self.embedder_config = get_default_embedder_config()
             self.cross_encoder_config = CrossEncoderConfig()
             self.data_config = DataConfig()
             self.transformer_config = HFModelConfig()

src/autointent/_wrappers/embedder/embedder.py

@@ -55,7 +55,6 @@ def _load_model(self) -> BaseEmbeddingBackend:
             return OpenaiEmbeddingBackend(self.config)
         # Check if it's exactly the abstract base config (not a subclass)
         if type(self.config) is EmbedderConfig:
-            # Handle abstract base config case
             msg = f"Cannot instantiate abstract EmbedderConfig: {self.config.__repr__()}"
             raise TypeError(msg)
         assert_never(self.config)

src/autointent/configs/__init__.py

@@ -1,6 +1,12 @@
 """Dataclasses for the configuration of the :class:`autointent.Embedder` and other objects."""
 
-from ._embedder import EmbedderConfig, OpenaiEmbeddingConfig, SentenceTransformerEmbeddingConfig, TaskTypeEnum
+from ._embedder import (
+    EmbedderConfig,
+    OpenaiEmbeddingConfig,
+    SentenceTransformerEmbeddingConfig,
+    TaskTypeEnum,
+    get_default_embedder_config,
+)
 from ._inference_node import InferenceNodeConfig
 from ._optimization import DataConfig, HPOConfig, LoggingConfig
 from ._torch import TorchTrainingConfig, VocabConfig
@@ -32,5 +38,6 @@
     "TorchTrainingConfig",
     "VectorIndexConfig",
     "VocabConfig",
+    "get_default_embedder_config",
     "get_default_vector_index_config",
 ]

src/autointent/configs/_embedder.py

@@ -96,3 +96,7 @@ class OpenaiEmbeddingConfig(EmbedderConfig):
     max_per_second: float | None = Field(
         None, description="Maximum number of API requests per second. Only used with async processing."
     )
+
+
+def get_default_embedder_config() -> EmbedderConfig:
+    return SentenceTransformerEmbeddingConfig()