Rename RecordText to EmbeddingInput

Why these changes are being introduced:

Code review suggested that 'RecordText' was a confusing name
for the object that we prepare to then create an embedding from.

How this addresses that need:

Renamign to 'EmbeddingInput' makes it crystal clear that we are
preparing an object that will be used to create an embedding.

Side effects of this change:
* None

Relevant ticket(s):
* https://mitlibraries.atlassian.net/browse/USE-112

Author: ghukill

embeddings/cli.py

@@ -230,10 +230,10 @@ def create_embeddings(
     # DEBUG ------------------------------------------------------------------------------
     import json  # noqa: PLC0415
 
-    from embeddings.embedding import RecordText  # noqa: PLC0415
+    from embeddings.embedding import EmbeddingInput  # noqa: PLC0415
 
     input_records = (
-        RecordText(
+        EmbeddingInput(
             timdex_record_id=timdex_record["timdex_record_id"],
             run_id=timdex_record["run_id"],
             run_record_offset=timdex_record["run_record_offset"],

embeddings/embedding.py

@@ -4,8 +4,12 @@
 
 
 @dataclass
-class RecordText:
-    """Input record for creating an embedding for.
+class EmbeddingInput:
+    """Encapsulates the inputs for an embedding.
+
+    When creating an embedding, we need to note what TIMDEX record the embedding is
+    associated with and what strategy was used to prepare the embedding input text from
+    the record itself.
 
     Args:
         (timdex_record_id, run_id, run_record_offset): composite key for TIMDEX record

embeddings/models/base.py

@@ -4,7 +4,7 @@
 from collections.abc import Iterator
 from pathlib import Path
 
-from embeddings.embedding import Embedding, RecordText
+from embeddings.embedding import Embedding, EmbeddingInput
 
 
 class BaseEmbeddingModel(ABC):
@@ -51,15 +51,15 @@ def load(self) -> None:
         """Load model from self.model_path."""
 
     @abstractmethod
-    def create_embedding(self, input_record: RecordText) -> Embedding:
-        """Create an Embedding for an RecordText.
+    def create_embedding(self, input_record: EmbeddingInput) -> Embedding:
+        """Create an Embedding for an EmbeddingInput.
 
         Args:
-            input_record: RecordText instance
+            input_record: EmbeddingInput instance
         """
 
     def create_embeddings(
-        self, input_records: Iterator[RecordText]
+        self, input_records: Iterator[EmbeddingInput]
     ) -> Iterator[Embedding]:
         """Yield Embeddings for an iterator of InputRecords.
 

embeddings/models/os_neural_sparse_doc_v3_gte.py

@@ -11,7 +11,7 @@
 from huggingface_hub import snapshot_download
 from transformers import AutoModelForMaskedLM, AutoTokenizer
 
-from embeddings.embedding import Embedding, RecordText
+from embeddings.embedding import Embedding, EmbeddingInput
 from embeddings.models.base import BaseEmbeddingModel
 
 if TYPE_CHECKING:
@@ -163,5 +163,5 @@ def load(self) -> None:
 
         logger.info(f"Model loaded successfully, {time.perf_counter()-start_time}s")
 
-    def create_embedding(self, input_record: RecordText) -> Embedding:
+    def create_embedding(self, input_record: EmbeddingInput) -> Embedding:
         raise NotImplementedError

tests/conftest.py

@@ -6,7 +6,7 @@
 import pytest
 from click.testing import CliRunner
 
-from embeddings.embedding import Embedding, RecordText
+from embeddings.embedding import Embedding, EmbeddingInput
 from embeddings.models import registry
 from embeddings.models.base import BaseEmbeddingModel
 
@@ -45,7 +45,7 @@ def download(self) -> Path:
     def load(self) -> None:
         logger.info("Model loaded successfully, 1.5s")
 
-    def create_embedding(self, input_record: RecordText) -> Embedding:
+    def create_embedding(self, input_record: EmbeddingInput) -> Embedding:
         return Embedding(
             timdex_record_id=input_record.timdex_record_id,
             run_id=input_record.run_id,

tests/test_models.py

@@ -2,7 +2,7 @@
 
 import pytest
 
-from embeddings.embedding import RecordText
+from embeddings.embedding import EmbeddingInput
 from embeddings.models.base import BaseEmbeddingModel
 from embeddings.models.registry import MODEL_REGISTRY, get_model_class
 
@@ -35,7 +35,7 @@ def test_mock_model_load(caplog, mock_model):
 
 
 def test_mock_model_create_embedding(mock_model):
-    input_record = RecordText(
+    input_record = EmbeddingInput(
         timdex_record_id="test-id",
         run_id="test-run",
         run_record_offset=42,
@@ -87,14 +87,14 @@ class InvalidModel(BaseEmbeddingModel):
 
 def test_base_model_create_embeddings_calls_create_embedding(mock_model):
     input_records = [
-        RecordText(
+        EmbeddingInput(
             timdex_record_id="id-1",
             run_id="run-1",
             run_record_offset=0,
             embedding_strategy="full_record",
             text="text 1",
         ),
-        RecordText(
+        EmbeddingInput(
             timdex_record_id="id-2",
             run_id="run-1",
             run_record_offset=1,