openaivec 0.14.13__py3-none-any.whl → 0.15.0__py3-none-any.whl

openaivec/pandas_ext.py

@@ -10,29 +10,32 @@ from openaivec import pandas_ext
 # (AZURE_OPENAI_API_KEY, AZURE_OPENAI_BASE_URL, AZURE_OPENAI_API_VERSION)
 # No explicit setup needed - clients are automatically created
 
-# Option 2: Use an existing OpenAI client instance
+# Option 2: Register an existing OpenAI client instance
 client = OpenAI(api_key="your-api-key")
-pandas_ext.use(client)
+pandas_ext.set_client(client)
 
-# Option 3: Use an existing Azure OpenAI client instance
+# Option 3: Register an Azure OpenAI client instance
 azure_client = AzureOpenAI(
     api_key="your-azure-key",
     base_url="https://YOUR-RESOURCE-NAME.services.ai.azure.com/openai/v1/",
     api_version="preview"
 )
-pandas_ext.use(azure_client)
+pandas_ext.set_client(azure_client)
 
-# Option 4: Use async Azure OpenAI client instance
+# Option 4: Register an async Azure OpenAI client instance
 async_azure_client = AsyncAzureOpenAI(
     api_key="your-azure-key",
     base_url="https://YOUR-RESOURCE-NAME.services.ai.azure.com/openai/v1/",
     api_version="preview"
 )
-pandas_ext.use_async(async_azure_client)
+pandas_ext.set_async_client(async_azure_client)
 
 # Set up model names (optional, defaults shown)
-pandas_ext.responses_model("gpt-4.1-mini")
-pandas_ext.embeddings_model("text-embedding-3-small")
+pandas_ext.set_responses_model("gpt-4.1-mini")
+pandas_ext.set_embeddings_model("text-embedding-3-small")
+
+# Inspect current configuration
+configured_model = pandas_ext.get_responses_model()
 ```
 
 This module provides `.ai` and `.aio` accessors for pandas Series and DataFrames
@@ -49,15 +52,6 @@ import numpy as np
 import pandas as pd
 import tiktoken
 from openai import AsyncOpenAI, OpenAI
-
-from openaivec._schema import InferredSchema, SchemaInferenceInput, SchemaInferer
-
-__all__ = [
-    "embeddings_model",
-    "responses_model",
-    "use",
-    "use_async",
-]
 from pydantic import BaseModel
 
 from openaivec._embeddings import AsyncBatchEmbeddings, BatchEmbeddings
@@ -65,13 +59,18 @@ from openaivec._model import EmbeddingsModelName, PreparedTask, ResponseFormat,
 from openaivec._provider import CONTAINER, _check_azure_v1_api_url
 from openaivec._proxy import AsyncBatchingMapProxy, BatchingMapProxy
 from openaivec._responses import AsyncBatchResponses, BatchResponses
+from openaivec._schema import InferredSchema, SchemaInferenceInput, SchemaInferer
 from openaivec.task.table import FillNaResponse, fillna
 
 __all__ = [
-    "use",
-    "use_async",
-    "responses_model",
-    "embeddings_model",
+    "get_async_client",
+    "get_client",
+    "get_embeddings_model",
+    "get_responses_model",
+    "set_async_client",
+    "set_client",
+    "set_embeddings_model",
+    "set_responses_model",
 ]
 
 _LOGGER = logging.getLogger(__name__)
@@ -95,37 +94,51 @@ def _df_rows_to_json_series(df: pd.DataFrame) -> pd.Series:
 T = TypeVar("T")  # For pipe function return type
 
 
-def use(client: OpenAI) -> None:
-    """Register a custom OpenAI‑compatible client.
+def set_client(client: OpenAI) -> None:
+    """Register a custom OpenAI-compatible client for pandas helpers.
 
     Args:
-        client (OpenAI): A pre‑configured `openai.OpenAI` or
-            `openai.AzureOpenAI` instance.
-            The same instance is reused by every helper in this module.
+        client (OpenAI): A pre-configured `openai.OpenAI` or
+            `openai.AzureOpenAI` instance reused by every helper in this module.
     """
-    # Check Azure v1 API URL if using AzureOpenAI client
     if client.__class__.__name__ == "AzureOpenAI" and hasattr(client, "base_url"):
         _check_azure_v1_api_url(str(client.base_url))
 
     CONTAINER.register(OpenAI, lambda: client)
 
 
-def use_async(client: AsyncOpenAI) -> None:
-    """Register a custom asynchronous OpenAI‑compatible client.
+def get_client() -> OpenAI:
+    """Get the currently registered OpenAI-compatible client.
+
+    Returns:
+        OpenAI: The registered `openai.OpenAI` or `openai.AzureOpenAI` instance.
+    """
+    return CONTAINER.resolve(OpenAI)
+
+
+def set_async_client(client: AsyncOpenAI) -> None:
+    """Register a custom asynchronous OpenAI-compatible client.
 
     Args:
-        client (AsyncOpenAI): A pre‑configured `openai.AsyncOpenAI` or
-            `openai.AsyncAzureOpenAI` instance.
-            The same instance is reused by every helper in this module.
+        client (AsyncOpenAI): A pre-configured `openai.AsyncOpenAI` or
+            `openai.AsyncAzureOpenAI` instance reused by every helper in this module.
     """
-    # Check Azure v1 API URL if using AsyncAzureOpenAI client
     if client.__class__.__name__ == "AsyncAzureOpenAI" and hasattr(client, "base_url"):
         _check_azure_v1_api_url(str(client.base_url))
 
     CONTAINER.register(AsyncOpenAI, lambda: client)
 
 
-def responses_model(name: str) -> None:
+def get_async_client() -> AsyncOpenAI:
+    """Get the currently registered asynchronous OpenAI-compatible client.
+
+    Returns:
+        AsyncOpenAI: The registered `openai.AsyncOpenAI` or `openai.AsyncAzureOpenAI` instance.
+    """
+    return CONTAINER.resolve(AsyncOpenAI)
+
+
+def set_responses_model(name: str) -> None:
     """Override the model used for text responses.
 
     Args:
@@ -135,7 +148,16 @@ def responses_model(name: str) -> None:
     CONTAINER.register(ResponsesModelName, lambda: ResponsesModelName(name))
 
 
-def embeddings_model(name: str) -> None:
+def get_responses_model() -> str:
+    """Get the currently registered model name for text responses.
+
+    Returns:
+        str: The model name (for example, ``gpt-4.1-mini``).
+    """
+    return CONTAINER.resolve(ResponsesModelName).value
+
+
+def set_embeddings_model(name: str) -> None:
     """Override the model used for text embeddings.
 
     Args:
@@ -145,6 +167,15 @@ def embeddings_model(name: str) -> None:
     CONTAINER.register(EmbeddingsModelName, lambda: EmbeddingsModelName(name))
 
 
+def get_embeddings_model() -> str:
+    """Get the currently registered model name for text embeddings.
+
+    Returns:
+        str: The model name (for example, ``text-embedding-3-small``).
+    """
+    return CONTAINER.resolve(EmbeddingsModelName).value
+
+
 def _extract_value(x, series_name):
     """Return a homogeneous ``dict`` representation of any Series value.
 
@@ -639,7 +670,7 @@ class OpenAIVecSeriesAccessor:
             animals.ai.count_tokens()
             ```
             This method uses the `tiktoken` library to count tokens based on the
-            model name set by `responses_model`.
+            model name configured via `set_responses_model`.
 
         Returns:
             pandas.Series: Token counts for each element.

openaivec/spark.py

@@ -193,8 +193,6 @@ def setup(
         CONTAINER.register(ResponsesModelName, lambda: ResponsesModelName(responses_model_name))
 
     if embeddings_model_name:
-        from openaivec._model import EmbeddingsModelName
-
         CONTAINER.register(EmbeddingsModelName, lambda: EmbeddingsModelName(embeddings_model_name))
 
     CONTAINER.clear_singletons()
@@ -244,6 +242,50 @@ def setup_azure(
     CONTAINER.clear_singletons()
 
 
+def set_responses_model(model_name: str):
+    """Set the default model name for response generation in the DI container.
+
+    Args:
+        model_name (str): The model name to set as default for responses.
+    """
+    CONTAINER.register(ResponsesModelName, lambda: ResponsesModelName(model_name))
+    CONTAINER.clear_singletons()
+
+
+def get_responses_model() -> str | None:
+    """Get the default model name for response generation from the DI container.
+
+    Returns:
+        str | None: The default model name for responses, or None if not set.
+    """
+    try:
+        return CONTAINER.resolve(ResponsesModelName).value
+    except Exception:
+        return None
+
+
+def set_embeddings_model(model_name: str):
+    """Set the default model name for embeddings in the DI container.
+
+    Args:
+        model_name (str): The model name to set as default for embeddings.
+    """
+    CONTAINER.register(EmbeddingsModelName, lambda: EmbeddingsModelName(model_name))
+    CONTAINER.clear_singletons()
+
+
+def get_embeddings_model() -> str | None:
+    """Get the default model name for embeddings from the DI container.
+
+    Returns:
+        str | None: The default model name for embeddings, or None if not set.
+    """
+    try:
+        return CONTAINER.resolve(EmbeddingsModelName).value
+    except Exception:
+        return None
+
+
 def _python_type_to_spark(python_type):
     origin = get_origin(python_type)
 
@@ -322,7 +364,7 @@ def _safe_dump(x: BaseModel | None) -> dict:
 def responses_udf(
     instructions: str,
     response_format: type[ResponseFormat] = str,
-    model_name: str = CONTAINER.resolve(ResponsesModelName).value,
+    model_name: str | None = None,
     batch_size: int | None = None,
     max_concurrency: int = 8,
     **api_kwargs,
@@ -351,8 +393,9 @@ def responses_udf(
         instructions (str): The system prompt or instructions for the model.
         response_format (type[ResponseFormat]): The desired output format. Either `str` for plain text
             or a Pydantic `BaseModel` for structured JSON output. Defaults to `str`.
-        model_name (str): For Azure OpenAI, use your deployment name (e.g., "my-gpt4-deployment").
-            For OpenAI, use the model name (e.g., "gpt-4.1-mini"). Defaults to configured model in DI container.
+        model_name (str | None): For Azure OpenAI, use your deployment name (e.g., "my-gpt4-deployment").
+            For OpenAI, use the model name (e.g., "gpt-4.1-mini"). Defaults to configured model in DI container
+            via ResponsesModelName if not provided.
         batch_size (int | None): Number of rows per async batch request within each partition.
             Larger values reduce API call overhead but increase memory usage.
             Defaults to None (automatic batch size optimization that dynamically
@@ -382,13 +425,15 @@ def responses_udf(
         - Consider your OpenAI tier limits: total_requests = max_concurrency × executors
         - Use Spark UI to optimize partition sizes relative to batch_size
     """
+    _model_name = model_name or CONTAINER.resolve(ResponsesModelName).value
+
     if issubclass(response_format, BaseModel):
         spark_schema = _pydantic_to_spark_schema(response_format)
         json_schema_string = serialize_base_model(response_format)
 
         @pandas_udf(returnType=spark_schema)  # type: ignore[call-overload]
         def structure_udf(col: Iterator[pd.Series]) -> Iterator[pd.DataFrame]:
-            pandas_ext.responses_model(model_name)
+            pandas_ext.set_responses_model(_model_name)
             response_format = deserialize_base_model(json_schema_string)
             cache = AsyncBatchingMapProxy[str, response_format](
                 batch_size=batch_size,
@@ -415,7 +460,7 @@ def responses_udf(
 
         @pandas_udf(returnType=StringType())  # type: ignore[call-overload]
         def string_udf(col: Iterator[pd.Series]) -> Iterator[pd.Series]:
-            pandas_ext.responses_model(model_name)
+            pandas_ext.set_responses_model(_model_name)
             cache = AsyncBatchingMapProxy[str, str](
                 batch_size=batch_size,
                 max_concurrency=max_concurrency,
@@ -443,7 +488,7 @@ def responses_udf(
 
 def task_udf(
     task: PreparedTask[ResponseFormat],
-    model_name: str = CONTAINER.resolve(ResponsesModelName).value,
+    model_name: str | None = None,
     batch_size: int | None = None,
     max_concurrency: int = 8,
     **api_kwargs,
@@ -459,8 +504,9 @@ def task_udf(
     Args:
         task (PreparedTask): A predefined task configuration containing instructions,
             response format, and API parameters.
-        model_name (str): For Azure OpenAI, use your deployment name (e.g., "my-gpt4-deployment").
-            For OpenAI, use the model name (e.g., "gpt-4.1-mini"). Defaults to configured model in DI container.
+        model_name (str | None): For Azure OpenAI, use your deployment name (e.g., "my-gpt4-deployment").
+            For OpenAI, use the model name (e.g., "gpt-4.1-mini"). Defaults to configured model in DI container
+            via ResponsesModelName if not provided.
         batch_size (int | None): Number of rows per async batch request within each partition.
             Larger values reduce API call overhead but increase memory usage.
             Defaults to None (automatic batch size optimization that dynamically
@@ -550,7 +596,7 @@ def parse_udf(
     example_table_name: str | None = None,
     example_field_name: str | None = None,
     max_examples: int = 100,
-    model_name: str = CONTAINER.resolve(ResponsesModelName).value,
+    model_name: str | None = None,
     batch_size: int | None = None,
     max_concurrency: int = 8,
     **api_kwargs,
@@ -574,8 +620,9 @@ def parse_udf(
             If provided, `example_table_name` must also be specified.
         max_examples (int): Maximum number of examples to retrieve for schema inference.
             Defaults to 100.
-        model_name (str): For Azure OpenAI, use your deployment name (e.g., "my-gpt4-deployment").
-            For OpenAI, use the model name (e.g., "gpt-4.1-mini"). Defaults to configured model in DI container.
+        model_name (str | None): For Azure OpenAI, use your deployment name (e.g., "my-gpt4-deployment").
+            For OpenAI, use the model name (e.g., "gpt-4.1-mini"). Defaults to configured model in DI container
+            via ResponsesModelName if not provided.
         batch_size (int | None): Number of rows per async batch request within each partition.
             Larger values reduce API call overhead but increase memory usage.
             Defaults to None (automatic batch size optimization that dynamically
@@ -622,7 +669,7 @@ def parse_udf(
 
 
 def embeddings_udf(
-    model_name: str = CONTAINER.resolve(EmbeddingsModelName).value,
+    model_name: str | None = None,
     batch_size: int | None = None,
     max_concurrency: int = 8,
     **api_kwargs,
@@ -648,9 +695,9 @@ def embeddings_udf(
             sc.environment["AZURE_OPENAI_API_VERSION"] = "preview"
 
     Args:
-        model_name (str): For Azure OpenAI, use your deployment name (e.g., "my-embedding-deployment").
+        model_name (str | None): For Azure OpenAI, use your deployment name (e.g., "my-embedding-deployment").
             For OpenAI, use the model name (e.g., "text-embedding-3-small").
-            Defaults to configured model in DI container.
+            Defaults to configured model in DI container via EmbeddingsModelName if not provided.
         batch_size (int | None): Number of rows per async batch request within each partition.
             Larger values reduce API call overhead but increase memory usage.
             Defaults to None (automatic batch size optimization that dynamically
@@ -678,9 +725,11 @@ def embeddings_udf(
         - Use larger batch_size for embeddings compared to response generation
     """
 
+    _model_name = model_name or CONTAINER.resolve(EmbeddingsModelName).value
+
     @pandas_udf(returnType=ArrayType(FloatType()))  # type: ignore[call-overload,misc]
     def _embeddings_udf(col: Iterator[pd.Series]) -> Iterator[pd.Series]:
-        pandas_ext.embeddings_model(model_name)
+        pandas_ext.set_embeddings_model(_model_name)
         cache = AsyncBatchingMapProxy[str, np.ndarray](
             batch_size=batch_size,
             max_concurrency=max_concurrency,

{openaivec-0.14.13.dist-info → openaivec-0.15.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: openaivec
-Version: 0.14.13
+Version: 0.15.0
 Summary: Generative mutation for tabular calculation
 Project-URL: Homepage, https://microsoft.github.io/openaivec/
 Project-URL: Repository, https://github.com/microsoft/openaivec
@@ -26,6 +26,8 @@ Description-Content-Type: text/markdown
 
 # openaivec
 
+[Contributor guidelines](AGENTS.md)
+
 **Transform your data analysis with AI-powered text processing at scale.**
 
 **openaivec** enables data analysts to seamlessly integrate OpenAI's language models into their pandas and Spark workflows. Process thousands of text records with natural language instructions, turning unstructured data into actionable insights with just a few lines of code.
@@ -187,13 +189,13 @@ os.environ["OPENAI_API_KEY"] = "your-api-key-here"
 
 # Authentication Option 2: Custom client (optional)
 # from openai import OpenAI, AsyncOpenAI
-# pandas_ext.use(OpenAI())
+# pandas_ext.set_client(OpenAI())
 # For async operations:
-# pandas_ext.use_async(AsyncOpenAI())
+# pandas_ext.set_async_client(AsyncOpenAI())
 
 # Configure model (optional - defaults to gpt-4.1-mini)
 # For Azure OpenAI: use your deployment name, for OpenAI: use model name
-pandas_ext.responses_model("gpt-4.1-mini")
+pandas_ext.set_responses_model("gpt-4.1-mini")
 
 # Create your data
 df = pd.DataFrame({"name": ["panda", "rabbit", "koala"]})
@@ -220,7 +222,7 @@ When using reasoning models (o1-preview, o1-mini, o3-mini, etc.), you must set `
 
 ```python
 # For reasoning models like o1-preview, o1-mini, o3-mini
-pandas_ext.responses_model("o1-mini")  # Set your reasoning model
+pandas_ext.set_responses_model("o1-mini")  # Set your reasoning model
 
 # MUST use temperature=None with reasoning models
 result = df.assign(
@@ -291,7 +293,7 @@ import pandas as pd
 from openaivec import pandas_ext
 
 # Setup (same as synchronous version)
-pandas_ext.responses_model("gpt-4.1-mini")
+pandas_ext.set_responses_model("gpt-4.1-mini")
 
 df = pd.DataFrame({"text": [
     "This product is amazing!",

{openaivec-0.14.13.dist-info → openaivec-0.15.0.dist-info}/RECORD

@@ -12,8 +12,8 @@ openaivec/_responses.py,sha256=qBrYv4qblDIs5dRvj9t96r8UfAJmy4ZvtAe6csNZ7oM,20412
 openaivec/_schema.py,sha256=iOeR5J_ihZRDZtzmqvOK1ZtInKcx4OnoR38DB3VmmQw,15666
 openaivec/_serialize.py,sha256=u2Om94Sc_QgJkTlW2BAGw8wd6gYDhc6IRqvS-qevFSs,8399
 openaivec/_util.py,sha256=XfueAycVCQvgRLS7wF7e306b53lebORvZOBzbQjy4vE,6438
-openaivec/pandas_ext.py,sha256=r2jpFqDnWcQYK3pMv5hCtOStOMltccDyLkpprLmIOls,85715
-openaivec/spark.py,sha256=zaEivVOe3ukG8coa9JEUyISQ1YcMqCvAbhaarvn2SOM,32507
+openaivec/pandas_ext.py,sha256=1euz52rwKpUBvWRCKvkDjwCFf_zNYnf60wF5OXHiCqw,86727
+openaivec/spark.py,sha256=8-Hap36D0kcyV8RMA-PyFjZxfAnMfgtcp9gKASRnUwU,34032
 openaivec/task/__init__.py,sha256=RkYIKrcE83M_9Um9cSMkeGzL9kPRAovajfRvr31YxLE,6178
 openaivec/task/customer_support/__init__.py,sha256=KWfGyXPdZyfGdRH17x7hPpJJ1N2EP9PPhZx0fvBAwSI,884
 openaivec/task/customer_support/customer_sentiment.py,sha256=d8spZUtImjePK0xWGvIW98ghbdyOZ0KEZmaUpG8QB7M,7532
@@ -31,7 +31,7 @@ openaivec/task/nlp/sentiment_analysis.py,sha256=u-zpqAaQYcr7I3mqMv_CTJXkfxtoLft3
 openaivec/task/nlp/translation.py,sha256=kgWj2oN8pUId3vuHTJNx636gB49AGEKXWICA_XJgE_0,6628
 openaivec/task/table/__init__.py,sha256=kJz15WDJXjyC7UIHKBvlTRhCf347PCDMH5T5fONV2sU,83
 openaivec/task/table/fillna.py,sha256=zL6m5hGD4kamV7qHETnn__B59wIY540Ks0EzNgUJgdI,6888
-openaivec-0.14.13.dist-info/METADATA,sha256=rB_WJhIVX11WUoA-r2Ryn57QIuTWj0q0JhjPlz6wXv4,28216
-openaivec-0.14.13.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-openaivec-0.14.13.dist-info/licenses/LICENSE,sha256=ws_MuBL-SCEBqPBFl9_FqZkaaydIJmxHrJG2parhU4M,1141
-openaivec-0.14.13.dist-info/RECORD,,
+openaivec-0.15.0.dist-info/METADATA,sha256=cVTYsT6TOMij_vagDgsIbo886U24Ys5dkah7ZvdEkdw,28278
+openaivec-0.15.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+openaivec-0.15.0.dist-info/licenses/LICENSE,sha256=ws_MuBL-SCEBqPBFl9_FqZkaaydIJmxHrJG2parhU4M,1141
+openaivec-0.15.0.dist-info/RECORD,,