feat(python): support optional vector field in pydantic model (#1097)

The LanceDB embeddings registry allows users to annotate the pydantic
model used as table schema with the desired embedding function, e.g.:

```python
class Schema(LanceModel):
    id: str
    vector: Vector(openai.ndims()) = openai.VectorField()
    text: str = openai.SourceField()
```

Tables created like this does not require embeddings to be calculated by
the user explicitly, e.g. this works:

```python
table.add([{"id": "foo", "text": "rust all the things"}])
```

However, trying to construct pydantic model instances without vector
doesn't because it's a required field.

Instead, you need add a default value:

```python
class Schema(LanceModel):
    id: str
    vector: Vector(openai.ndims()) = openai.VectorField(default=None)
    text: str = openai.SourceField()
```

then this completes without errors:
```python
table.add([Schema(id="foo", text="rust all the things")])
```

However, all of the vectors are filled with zeros. Instead in
add_vector_col we have to add an additional check so that the embedding
generation is called.

python/python/lancedb/table.py

@@ -117,7 +117,8 @@ def _append_vector_col(data: pa.Table, metadata: dict, schema: Optional[pa.Schem
     functions = EmbeddingFunctionRegistry.get_instance().parse_functions(metadata)
     for vector_column, conf in functions.items():
         func = conf.function
-        if vector_column not in data.column_names:
+        no_vector_column = vector_column not in data.column_names
+        if no_vector_column or pc.all(pc.is_null(data[vector_column])).as_py():
             col_data = func.compute_source_embeddings_with_retry(
                 data[conf.source_column]
             )
@@ -125,9 +126,16 @@ def _append_vector_col(data: pa.Table, metadata: dict, schema: Optional[pa.Schem
                 dtype = schema.field(vector_column).type
             else:
                 dtype = pa.list_(pa.float32(), len(col_data[0]))
-            data = data.append_column(
-                pa.field(vector_column, type=dtype), pa.array(col_data, type=dtype)
-            )
+            if no_vector_column:
+                data = data.append_column(
+                    pa.field(vector_column, type=dtype), pa.array(col_data, type=dtype)
+                )
+            else:
+                data = data.set_column(
+                    data.column_names.index(vector_column),
+                    pa.field(vector_column, type=dtype),
+                    pa.array(col_data, type=dtype),
+                )
     return data
 
 

python/python/tests/test_embeddings.py

@@ -11,6 +11,7 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 import sys
+from typing import List, Union
 
 import lance
 import lancedb
@@ -23,6 +24,8 @@ from lancedb.embeddings import (
     EmbeddingFunctionRegistry,
     with_embeddings,
 )
+from lancedb.embeddings.base import TextEmbeddingFunction
+from lancedb.embeddings.registry import get_registry, register
 from lancedb.pydantic import LanceModel, Vector
 
 
@@ -112,3 +115,34 @@ def test_embedding_function_rate_limit(tmp_path):
     table.add([{"text": "hello world"}])
     table.add([{"text": "hello world"}])
     assert len(table) == 2
+
+
+def test_add_optional_vector(tmp_path):
+    @register("mock-embedding")
+    class MockEmbeddingFunction(TextEmbeddingFunction):
+        def ndims(self):
+            return 128
+
+        def generate_embeddings(
+            self, texts: Union[List[str], np.ndarray]
+        ) -> List[np.array]:
+            """
+            Generate the embeddings for the given texts
+            """
+            return [np.random.randn(self.ndims()).tolist() for _ in range(len(texts))]
+
+    registry = get_registry()
+    model = registry.get("mock-embedding").create()
+
+    class LanceSchema(LanceModel):
+        id: str
+        vector: Vector(model.ndims()) = model.VectorField(default=None)
+        text: str = model.SourceField()
+
+    db = lancedb.connect(tmp_path)
+    tbl = db.create_table("optional_vector", schema=LanceSchema)
+
+    # add works
+    expected = LanceSchema(id="id", text="text")
+    tbl.add([expected])
+    assert not (np.abs(tbl.to_pandas()["vector"][0]) < 1e-6).all()