snowflake-ml-python 1.28.0__tar.gz → 1.30.0__tar.gz

{snowflake_ml_python-1.28.0 → snowflake_ml_python-1.30.0}/CHANGELOG.md

@@ -1,6 +1,55 @@
 # Release History
 
-## 1.28.0
+## 1.30.0
+
+### New Features
+
+* Experiment Tracking live logging (PrPr): In SPCS, call `set_live_logging_status(True)` to automatically capture and
+  persist outputs to stdout and stderr while a run is active. The captured logs can be viewed from the Experiments UI.
+* Registry: Support logging MLflow models created via `mlflow.*.save_model()` to the Snowflake Model
+  Registry. Previously, only models logged through `mlflow.*.log_model()` were supported. This also
+  enables logging custom `mlflow.pyfunc.PythonModel` subclasses saved locally.
+
+### Bug Fixes
+
+* Registry: Fixed Prophet model handler to correctly mark the `predict` method as partitioned, ensuring it
+  uses a partitioned `TABLE_FUNCTION` when deployed to Snowflake.
+* Registry: Support `text-generation` models without chat template. The model will have signatures to automatically take
+  plain strings as input without needing to specify the signatures in `log_model`.
+  The signature can be overridden if the user chooses to.
+
+### Behavior Changes
+
+* Registry: Huggingface models with task `text-generation` that do not have chat templates will be logged with signature
+  that supports plain text (string) as input.
+
+### Deprecations
+
+* Registry: Removed support for logging Hugging Face Pipelines in config-only mode. Config-only
+  models could not run in warehouse and required an External Access Integration (EAI) with egress
+  to Hugging Face hosts. Use remote logging or local download instead — these approaches support
+  warehouse execution and store model weights at log time, enabling fully air-gapped services.
+
+## 1.29.0 (2026-02-24)
+
+### New Features
+
+* Model serving: Introducing `InferenceEngine.PYTHON_GENERIC` enum value.
+  Users can pass `InferenceEngine.PYTHON_GENERIC` to use a Python-based inference server for model serving.
+
+### Bug Fixes
+
+* Registry: Fixed a bug where using inference parameters (ParamSpec) with table function or
+  partitioned model methods would fail at runtime with a `NameError`.
+* Registry: Fixed a bug where MLflow models with string columns failed during inference with
+  "Can not safely convert string to \<U0\>" errors due to MLflow's schema validation not handling
+  `pd.StringDtype` correctly.
+
+### Behavior Changes
+
+### Deprecations
+
+## 1.28.0 (2026-02-17)
 
 ### New Features
 

{snowflake_ml_python-1.28.0/snowflake_ml_python.egg-info → snowflake_ml_python-1.30.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: snowflake-ml-python
-Version: 1.28.0
+Version: 1.30.0
 Summary: The machine learning client library that is used for interacting with Snowflake to build machine learning solutions.
 Author-email: "Snowflake, Inc" <support@snowflake.com>
 License: 
@@ -255,7 +255,6 @@ Requires-Dist: scipy<2,>=1.9
 Requires-Dist: shap<1,>=0.46.0
 Requires-Dist: snowflake-connector-python[pandas]<5,>=3.17.3
 Requires-Dist: snowflake-snowpark-python!=1.26.0,<2,>=1.17.0
-Requires-Dist: snowflake.core<2,>=1.0.2
 Requires-Dist: sqlparse<1,>=0.4
 Requires-Dist: tqdm<5
 Requires-Dist: typing-extensions<5,>=4.1.0
@@ -417,7 +416,56 @@ NOTE: Version 1.7.0 is used as example here. Please choose the the latest versio
 
 # Release History
 
-## 1.28.0
+## 1.30.0
+
+### New Features
+
+* Experiment Tracking live logging (PrPr): In SPCS, call `set_live_logging_status(True)` to automatically capture and
+  persist outputs to stdout and stderr while a run is active. The captured logs can be viewed from the Experiments UI.
+* Registry: Support logging MLflow models created via `mlflow.*.save_model()` to the Snowflake Model
+  Registry. Previously, only models logged through `mlflow.*.log_model()` were supported. This also
+  enables logging custom `mlflow.pyfunc.PythonModel` subclasses saved locally.
+
+### Bug Fixes
+
+* Registry: Fixed Prophet model handler to correctly mark the `predict` method as partitioned, ensuring it
+  uses a partitioned `TABLE_FUNCTION` when deployed to Snowflake.
+* Registry: Support `text-generation` models without chat template. The model will have signatures to automatically take
+  plain strings as input without needing to specify the signatures in `log_model`.
+  The signature can be overridden if the user chooses to.
+
+### Behavior Changes
+
+* Registry: Huggingface models with task `text-generation` that do not have chat templates will be logged with signature
+  that supports plain text (string) as input.
+
+### Deprecations
+
+* Registry: Removed support for logging Hugging Face Pipelines in config-only mode. Config-only
+  models could not run in warehouse and required an External Access Integration (EAI) with egress
+  to Hugging Face hosts. Use remote logging or local download instead — these approaches support
+  warehouse execution and store model weights at log time, enabling fully air-gapped services.
+
+## 1.29.0 (2026-02-24)
+
+### New Features
+
+* Model serving: Introducing `InferenceEngine.PYTHON_GENERIC` enum value.
+  Users can pass `InferenceEngine.PYTHON_GENERIC` to use a Python-based inference server for model serving.
+
+### Bug Fixes
+
+* Registry: Fixed a bug where using inference parameters (ParamSpec) with table function or
+  partitioned model methods would fail at runtime with a `NameError`.
+* Registry: Fixed a bug where MLflow models with string columns failed during inference with
+  "Can not safely convert string to \<U0\>" errors due to MLflow's schema validation not handling
+  `pd.StringDtype` correctly.
+
+### Behavior Changes
+
+### Deprecations
+
+## 1.28.0 (2026-02-17)
 
 ### New Features
 

{snowflake_ml_python-1.28.0 → snowflake_ml_python-1.30.0}/pyproject.toml

@@ -8,7 +8,7 @@ description = "The machine learning client library that is used for interacting
 classifiers = [ "Development Status :: 5 - Production/Stable", "Environment :: Console", "Environment :: Other Environment", "Intended Audience :: Developers", "Intended Audience :: Education", "Intended Audience :: Information Technology", "Intended Audience :: System Administrators", "License :: OSI Approved :: Apache Software License", "Operating System :: OS Independent", "Programming Language :: Python :: 3.9", "Programming Language :: Python :: 3.10", "Programming Language :: Python :: 3.11", "Programming Language :: Python :: 3.12", "Topic :: Database", "Topic :: Software Development", "Topic :: Software Development :: Libraries", "Topic :: Software Development :: Libraries :: Application Frameworks", "Topic :: Software Development :: Libraries :: Python Modules", "Topic :: Scientific/Engineering :: Information Analysis",]
 requires-python = ">=3.9, <3.14"
 dynamic = [ "version", "readme",]
-dependencies = [ "anyio>=3.5.0,<5", "cachetools>=3.1.1,<6", "cloudpickle>=2.0.0", "cryptography", "fsspec[http]>=2024.6.1,<2026", "importlib_resources>=6.1.1, <7", "numpy>=1.23,<3", "packaging>=20.9,<26", "pandas>=2.1.4,<3", "platformdirs<5", "pyarrow<19.0.0", "pydantic>=2.8.2, <3", "pyjwt>=2.0.0, <3", "pytimeparse>=1.1.8,<2", "pyyaml>=6.0,<7", "retrying>=1.3.3,<2", "s3fs>=2024.6.1,<2026", "scikit-learn<1.8", "scipy>=1.9,<2", "shap>=0.46.0,<1", "snowflake-connector-python[pandas]>=3.17.3,<5", "snowflake-snowpark-python>=1.17.0,<2,!=1.26.0", "snowflake.core>=1.0.2,<2", "sqlparse>=0.4,<1", "tqdm<5", "typing-extensions>=4.1.0,<5", "xgboost<4",]
+dependencies = [ "anyio>=3.5.0,<5", "cachetools>=3.1.1,<6", "cloudpickle>=2.0.0", "cryptography", "fsspec[http]>=2024.6.1,<2026", "importlib_resources>=6.1.1, <7", "numpy>=1.23,<3", "packaging>=20.9,<26", "pandas>=2.1.4,<3", "platformdirs<5", "pyarrow<19.0.0", "pydantic>=2.8.2, <3", "pyjwt>=2.0.0, <3", "pytimeparse>=1.1.8,<2", "pyyaml>=6.0,<7", "retrying>=1.3.3,<2", "s3fs>=2024.6.1,<2026", "scikit-learn<1.8", "scipy>=1.9,<2", "shap>=0.46.0,<1", "snowflake-connector-python[pandas]>=3.17.3,<5", "snowflake-snowpark-python>=1.17.0,<2,!=1.26.0", "sqlparse>=0.4,<1", "tqdm<5", "typing-extensions>=4.1.0,<5", "xgboost<4",]
 [[project.authors]]
 name = "Snowflake, Inc"
 email = "support@snowflake.com"

{snowflake_ml_python-1.28.0 → snowflake_ml_python-1.30.0}/snowflake/cortex/_complete.py

@@ -7,7 +7,6 @@ from typing import Any, Callable, Iterator, Optional, TypedDict, Union, cast
 from urllib.parse import urlunparse
 
 import requests
-from snowflake.core.rest import RESTResponse
 from typing_extensions import NotRequired, deprecated
 
 from snowflake import snowpark
@@ -80,7 +79,7 @@ class MidStreamException(Exception):
     def __init__(
         self,
         reason: typing.Optional[str] = None,
-        http_resp: typing.Optional["RESTResponse"] = None,
+        http_resp: typing.Optional[Any] = None,
         request_id: typing.Optional[str] = None,
     ) -> None:
         message = ""

{snowflake_ml_python-1.28.0 → snowflake_ml_python-1.30.0}/snowflake/ml/_internal/file_utils.py

@@ -281,6 +281,38 @@ def _retry_on_sql_error(exception: Exception) -> bool:
     return isinstance(exception, snowpark_exceptions.SnowparkSQLException)
 
 
+def _resolve_stage_dir_path(
+    stage_path: Union[pathlib.PurePosixPath, parse.ParseResult],
+    relative_path: pathlib.PurePosixPath,
+) -> str:
+    """Resolve the stage directory path for a file upload.
+
+    Combines the base stage path with the relative file path and returns the
+    parent directory. Uses PurePosixPath for cross-platform compatibility
+    (PosixPath cannot be instantiated on Windows).
+
+    Args:
+        stage_path: Base stage path, either a PurePosixPath or a parsed URL.
+        relative_path: POSIX-style relative path of the file within the local directory.
+
+    Returns:
+        The resolved stage directory path as a string (or unparsed URL).
+    """
+    if isinstance(stage_path, parse.ParseResult):
+        relative_stage_path = (pathlib.PurePosixPath(stage_path.path) / relative_path).parent
+        new_url = parse.ParseResult(
+            scheme=stage_path.scheme,
+            netloc=stage_path.netloc,
+            path=str(relative_stage_path),
+            params=stage_path.params,
+            query=stage_path.query,
+            fragment=stage_path.fragment,
+        )
+        return parse.urlunparse(new_url)
+    else:
+        return str((stage_path / relative_path).parent)
+
+
 def upload_directory_to_stage(
     session: snowpark.Session,
     local_path: pathlib.Path,
@@ -305,20 +337,7 @@ def upload_directory_to_stage(
         for filename in filenames:
             local_file_path = root_path / filename
             relative_path = pathlib.PurePosixPath(local_file_path.relative_to(local_path).as_posix())
-
-            if isinstance(stage_path, parse.ParseResult):
-                relative_stage_path = (pathlib.PosixPath(stage_path.path) / relative_path).parent
-                new_url = parse.ParseResult(
-                    scheme=stage_path.scheme,
-                    netloc=stage_path.netloc,
-                    path=str(relative_stage_path),
-                    params=stage_path.params,
-                    query=stage_path.query,
-                    fragment=stage_path.fragment,
-                )
-                stage_dir_path = parse.urlunparse(new_url)
-            else:
-                stage_dir_path = str((stage_path / relative_path).parent)
+            stage_dir_path = _resolve_stage_dir_path(stage_path, relative_path)
 
             retrying.retry(
                 retry_on_exception=_retry_on_sql_error,

{snowflake_ml_python-1.28.0 → snowflake_ml_python-1.30.0}/snowflake/ml/_internal/lineage/lineage_utils.py

@@ -72,6 +72,7 @@ def patch_dataframe(
     # Instance-level monkey-patches
     funcs = [
         "_with_plan",
+        "_copy_without_ast",
         "_lateral",
         "group_by",
         "group_by_grouping_sets",

snowflake_ml_python-1.30.0/snowflake/ml/_internal/utils/tee.py

@@ -0,0 +1,85 @@
+import io
+from typing import Any, Iterable, Iterator, Optional, TextIO
+
+
+class OutputTee(TextIO):
+    """A class that duplicates string writes to multiple file-like objects."""
+
+    def __init__(self, *streams: TextIO) -> None:
+        """Initialize the OutputTee with a list of output streams."""
+        super().__init__()
+        self.streams = streams
+        if not self.writable():
+            raise ValueError("All inputs to OutputTee must be writable.")
+
+    def close(self) -> None:
+        """Does not do anything. It is implemented this way because some of the streams may be in use elsewhere.
+        It is the responsibility of the caller to close any streams that need to be closed."""
+        pass
+
+    def fileno(self) -> int:
+        raise io.UnsupportedOperation("OutputTee does not support fileno")
+
+    def flush(self) -> None:
+        """Flush all streams."""
+        for stream in self.streams:
+            stream.flush()
+
+    def isatty(self) -> bool:
+        return False
+
+    def read(self, n: int = -1) -> str:
+        raise io.UnsupportedOperation("OutputTee does not support reading")
+
+    def readable(self) -> bool:
+        """OutputTee is not readable."""
+        return False
+
+    def readline(self, limit: int = -1) -> str:
+        raise io.UnsupportedOperation("OutputTee does not support reading")
+
+    def readlines(self, hint: int = -1) -> list[str]:
+        raise io.UnsupportedOperation("OutputTee does not support reading")
+
+    def seek(self, offset: int, whence: int = 0) -> int:
+        raise io.UnsupportedOperation("OutputTee does not support seek")
+
+    def seekable(self) -> bool:
+        """OutputTee is not seekable."""
+        return False
+
+    def tell(self) -> int:
+        raise io.UnsupportedOperation("OutputTee does not support tell")
+
+    def truncate(self, size: Optional[int] = None) -> int:
+        raise io.UnsupportedOperation("OutputTee does not support truncate")
+
+    def writable(self) -> bool:
+        """OutputTee is writable if and only if all streams are writable."""
+        return all(stream.writable() for stream in self.streams)
+
+    def write(self, data: str) -> int:
+        """Write to all streams."""
+        for stream in self.streams:
+            stream.write(data)
+        return len(data)
+
+    def writelines(self, lines: Iterable[str]) -> None:
+        """Write lines to all streams."""
+        lines_list = list(lines)
+        for stream in self.streams:
+            stream.writelines(lines_list)
+
+    def __enter__(self) -> "OutputTee":
+        """Enter context manager."""
+        return self
+
+    def __exit__(self, exc_type: Any, exc_val: Any, exc_tb: Any) -> None:
+        """Exit context manager."""
+        self.close()
+
+    def __iter__(self) -> Iterator[str]:
+        raise io.UnsupportedOperation("OutputTee does not support reading")
+
+    def __next__(self) -> str:
+        raise io.UnsupportedOperation("OutputTee does not support reading")

{snowflake_ml_python-1.28.0 → snowflake_ml_python-1.30.0}/snowflake/ml/data/data_connector.py

@@ -29,7 +29,7 @@ if TYPE_CHECKING:
 
     # This module can't actually depend on dataset to avoid a circular dependency
     # Dataset -> DatasetReader -> DataConnector -!-> Dataset
-    from snowflake.ml import dataset
+    from snowflake.ml import dataset  # type: ignore[attr-defined, unused-ignore]
 
 _PROJECT = "DataConnector"
 

{snowflake_ml_python-1.28.0 → snowflake_ml_python-1.30.0}/snowflake/ml/experiment/_client/experiment_tracking_sql_client.py

@@ -40,6 +40,23 @@ class ExperimentTrackingSQLClient(_base._BaseSQLClient):
             self._session, f"CREATE EXPERIMENT {if_not_exists_sql} {experiment_fqn}"
         ).has_dimensions(expected_rows=1, expected_cols=1).validate()
 
+    @telemetry.send_api_usage_telemetry(project=telemetry.TelemetryProject.EXPERIMENT_TRACKING.value)
+    def get_experiment_id(
+        self,
+        experiment_name: sql_identifier.SqlIdentifier,
+    ) -> int:
+        experiment_fqn = self.fully_qualified_object_name(self._database_name, self._schema_name, experiment_name)
+
+        result = (
+            query_result_checker.SqlResultValidator(
+                self._session,
+                f"CALL SYSTEM$RESOLVE_EXPERIMENT_ID('{experiment_fqn}')",
+            )
+            .has_dimensions(expected_rows=1, expected_cols=1)
+            .validate()
+        )
+        return int(result[0][0])
+
     @telemetry.send_api_usage_telemetry(project=telemetry.TelemetryProject.EXPERIMENT_TRACKING.value)
     def drop_experiment(
         self,
@@ -72,6 +89,23 @@ class ExperimentTrackingSQLClient(_base._BaseSQLClient):
             self._session, f"ALTER EXPERIMENT {experiment_fqn} COMMIT RUN {run_name}"
         ).has_dimensions(expected_rows=1, expected_cols=1).validate()
 
+    def get_run_id(
+        self,
+        experiment_name: sql_identifier.SqlIdentifier,
+        run_name: sql_identifier.SqlIdentifier,
+    ) -> int:
+        experiment_fqn = self.fully_qualified_object_name(self._database_name, self._schema_name, experiment_name)
+
+        result = (
+            query_result_checker.SqlResultValidator(
+                self._session,
+                f"CALL SYSTEM$RESOLVE_EXPERIMENT_RUN_ID('{experiment_fqn}', '{run_name}')",
+            )
+            .has_dimensions(expected_rows=1, expected_cols=1)
+            .validate()
+        )
+        return int(result[0][0])
+
     @telemetry.send_api_usage_telemetry(project=telemetry.TelemetryProject.EXPERIMENT_TRACKING.value)
     def drop_run(
         self, *, experiment_name: sql_identifier.SqlIdentifier, run_name: sql_identifier.SqlIdentifier

snowflake_ml_python-1.30.0/snowflake/ml/experiment/_logging/__init__.py

@@ -0,0 +1,6 @@
+from snowflake.ml.experiment._logging.experiment_logger import ExperimentLogger
+from snowflake.ml.experiment._logging.experiment_logging_context import (
+    ExperimentLoggingContext,
+)
+
+__all__ = ["ExperimentLogger", "ExperimentLoggingContext"]

snowflake_ml_python-1.30.0/snowflake/ml/experiment/_logging/experiment_logger.py

@@ -0,0 +1,90 @@
+import io
+import json
+import os
+from typing import Any, Iterable, Iterator, TextIO
+
+
+class ExperimentLogger(TextIO):
+    OUTPUT_DIRECTORY = "/var/log/services/external/experiment_tracking"
+
+    def __init__(
+        self,
+        exp_id: int,
+        run_id: int,
+        stream: str,
+    ) -> None:
+        super().__init__()
+        filepath = os.path.join(self.OUTPUT_DIRECTORY, str(exp_id), str(run_id), f"{stream}.log")
+        os.makedirs(os.path.dirname(filepath), exist_ok=True)
+        self.file = open(filepath, "a", buffering=1)  # Line buffering
+        self.exp_id = exp_id
+        self.run_id = run_id
+        self.stream = stream
+
+    def close(self) -> None:
+        self.file.close()
+
+    def fileno(self) -> int:
+        return self.file.fileno()
+
+    def flush(self) -> None:
+        self.file.flush()
+
+    def isatty(self) -> bool:
+        return False
+
+    def readable(self) -> bool:
+        return False
+
+    def read(self, n: int = -1) -> str:
+        raise io.UnsupportedOperation("read")
+
+    def readline(self, limit: int = -1) -> str:
+        raise io.UnsupportedOperation("readline")
+
+    def readlines(self, hint: int = -1) -> list[str]:
+        raise io.UnsupportedOperation("readlines")
+
+    def seekable(self) -> bool:
+        return False
+
+    def seek(self, offset: int, whence: int = 0) -> int:
+        raise io.UnsupportedOperation("seek")
+
+    def tell(self) -> int:
+        raise io.UnsupportedOperation("tell")
+
+    def truncate(self, size: int | None = None) -> int:
+        raise io.UnsupportedOperation("truncate")
+
+    def writable(self) -> bool:
+        return True
+
+    def write(self, data: str) -> int:
+        log_message = {
+            "body": data,
+            "attributes": {
+                "snow.experiment.id": self.exp_id,
+                "snow.experiment.run.id": self.run_id,
+                "snow.experiment.stream": self.stream,
+            },
+        }
+        json_data = json.dumps(log_message)
+        self.file.write(json_data + "\n")
+        return len(data)  # Return length of input, not JSON output
+
+    def writelines(self, lines: Iterable[str]) -> None:
+        for line in lines:
+            self.write(line)
+
+    def __enter__(self) -> "ExperimentLogger":
+        return self
+
+    def __exit__(self, exc_type: Any, exc_val: Any, exc_tb: Any) -> None:
+        self.close()
+
+    def __iter__(self) -> Iterator[str]:
+        raise io.UnsupportedOperation("__iter__")
+
+    def __next__(self) -> str:
+        raise io.UnsupportedOperation("__next__")

snowflake_ml_python-1.30.0/snowflake/ml/experiment/_logging/experiment_logging_context.py

@@ -0,0 +1,13 @@
+import contextlib
+from dataclasses import dataclass
+
+from snowflake.ml._internal.utils import tee
+from snowflake.ml.experiment._logging.experiment_logger import ExperimentLogger
+
+
+@dataclass(frozen=True)
+class ExperimentLoggingContext:
+    stdout_logger: ExperimentLogger
+    stderr_logger: ExperimentLogger
+    stdout_ctx: contextlib.redirect_stdout[tee.OutputTee]
+    stderr_ctx: contextlib.redirect_stderr[tee.OutputTee]