1.7.2

1.7.2

Bug Fixes

• Model Explainability: Fix issue that explain is enabled for scikit-learn pipeline
  whose task is UNKNOWN and fails later when invoked.

Behavior Changes

New Features

• Registry: Support asynchronous model inference service creation with the block option
  in ModelVersion.create_service() set to True by default.

1.7.1

1.7.1

Bug Fixes

• Registry: Null value is now allowed in the dataframe used in model signature inference. Null values will be ignored
  and others will be used to infer the signature.
• Registry: Pandas Extension DTypes (pandas.StringDType(), pandas.BooleanDType(), etc.) are now supported in model
  signature inference.
• Registry: Null value is now allowed in the dataframe used to predict.
• Data: Fix missing snowflake.ml.data.* module exports in wheel
• Dataset: Fix missing snowflake.ml.dataset.* module exports in wheel.
• Registry: Fix the issue that tf_keras.Model is not recognized as keras model when logging.

Behavior Changes

New Features

• Registry: Option to enable_monitoring set to False by default. This will gate access to preview features of Model Monitoring.
• Model Monitoring: show_model_monitors Registry method. This feature is still in Private Preview.
• Registry: Support pd.Series in input and output data.
• Model Monitoring: add_monitor Registry method. This feature is still in Private Preview.
• Model Monitoring: resume and suspend ModelMonitor. This feature is still in Private Preview.
• Model Monitoring: get_monitor Registry method. This feature is still in Private Preview.
• Model Monitoring: delete_monitor Registry method. This feature is still in Private Preview.

1.7.0

1.7.0

Behavior Change

• Generic: Require python >= 3.9.
• Data Connector: Update to_torch_dataset and to_torch_datapipe to add a dimension for scalar data.
  This allows for more seamless integration with PyTorch DataLoader, which creates batches by stacking inputs of each batch.

Examples:

ds = connector.to_torch_dataset(shuffle=False, batch_size=3)

• Input: "col1": [10, 11, 12]

  • Previous batch: array([10., 11., 12.]) with shape (3,)
  • New batch: array([[10.], [11.], [12.]]) with shape (3, 1)

• Input: "col2": [[0, 100], [1, 110], [2, 200]]

  • Previous batch: array([[ 0, 100], [ 1, 110], [ 2, 200]]) with shape (3,2)
  • New batch: No change

• Model Registry: External access integrations are optional when creating a model inference service in
  Snowflake >= 8.40.0.

• Model Registry: Deprecate build_external_access_integration with build_external_access_integrations in
  ModelVersion.create_service().

Bug Fixes

• Registry: Updated log_model API to accept both signature and sample_input_data parameters.
• Feature Store: ExampleHelper uses fully qualified path for table name. change weather features aggregation from 1d to 1h.
• Data Connector: Return numpy array with appropriate object type instead of list for multi-dimensional
  data from to_torch_dataset and to_torch_datapipe
• Model explainability: Incompatibility between SHAP 0.42.1 and XGB 2.1.1 resolved by using latest SHAP 0.46.0.

New Features

• Registry: Provide pass keyworded variable length of arguments to class ModelContext. Example usage:

mc = custom_model.ModelContext(
    config = 'local_model_dir/config.json',
    m1 = model1
)

class ExamplePipelineModel(custom_model.CustomModel):
    def __init__(self, context: custom_model.ModelContext) -> None:
      super().__init__(context)
      v = open(self.context['config']).read()
      self.bias = json.loads(v)['bias']

    @custom_model.inference_api
    def predict(self, input: pd.DataFrame) -> pd.DataFrame:
      model_output = self.context['m1'].predict(input)
      return pd.DataFrame({'output': model_output + self.bias})

• Model Development: Upgrade scikit-learn in UDTF backend for log_loss metric. As a result, eps argument is now ignored.
• Data Connector: Add the option of passing a None sized batch to to_torch_dataset for better
  interoperability with PyTorch DataLoader.
• Model Registry: Support pandas.CategoricalDtype
• Registry: It is now possible to pass signatures and sample_input_data at the same time to capture background
  data from explainablity and data lineage.

1.6.4

1.6.4

Bug Fixes

• Registry: Fix an issue that leads to incident when using ModelVersion.run with service.

1.6.3

1.6.3

• Model Registry (PrPr) has been removed.

Bug Fixes

• Registry: Fix a bug that when package whose name does not follow PEP-508 is provided when logging the model,
  an unexpected normalization is happening.
• Registry: Fix not a valid remote uri error when logging mlflow models.
• Registry: Fix a bug that ModelVersion.run is called in a nested way.
• Registry: Fix an issue that leads to log_model failure when local package version contains parts other than
  base version.

New Features

• Data: Improve DataConnector.to_pandas() performance when loading from Snowpark DataFrames.
• Model Registry: Allow users to set a model task while using log_model.
• Feature Store: FeatureView supports ON_CREATE or ON_SCHEDULE initialize mode.

1.6.2

1.6.2 (TBD)

Bug Fixes

• Modeling: Support XGBoost version that is larger than 2.

• Data: Fix multiple epoch iteration over DataConnector.to_torch_datapipe() DataPipes.

• Generic: Fix a bug that when an invalid name is provided to argument where fully qualified name is expected, it will
  be parsed wrongly. Now it raises an exception correctly.

• Model Explainability: Handle explanations for multiclass XGBoost classification models

• Model Explainability: Workarounds and better error handling for XGB>2.1.0 not working with SHAP==0.42.1

New Features

• Data: Add top-level exports for DataConnector and DataSource to snowflake.ml.data.
• Data: Add native batching support via batch_size and drop_last_batch arguments to DataConnector.to_torch_dataset()
• Feature Store: update_feature_view() supports taking feature view object as argument.

Behavior Changes

1.6.1

1.6.1 (2024-08-12)

Bug Fixes

• Feature Store: Support large metadata blob when generating dataset
• Feature Store: Added a hidden knob in FeatureView as kargs for setting customized
  refresh_mode
• Registry: Fix an error message in Model Version run when function_name is not mentioned and model has multiple
  target methods.
• Cortex inference: snowflake.cortex.Complete now only uses the REST API for streaming and the use_rest_api_experimental
  is no longer needed.
• Feature Store: Add a new API: FeatureView.list_columns() which list all column information.
• Data: Fix DataFrame ingestion with ArrowIngestor.

New Features

• Enable set_params to set the parameters of the underlying sklearn estimator, if the snowflake-ml model has been fit.
• Data: Add top-level exports for DataConnector and DataSource to snowflake.ml.data.
• Data: Add snowflake.ml.data.ingestor_utils module with utility functions helpful for DataIngestor implementations.
• Data: Add new to_torch_dataset() connector to DataConnector to replace deprecated DataPipe.
• Registry: Option to enable_explainability set to True by default for XGBoost, LightGBM and CatBoost as PuPr feature.
• Registry: Option to enable_explainability when registering SHAP supported sklearn models.

Behavior Changes

1.6.0

1.6.0

Bug Fixes

• Modeling: SimpleImputer can impute integer columns with integer values.
• Registry: Fix an issue when providing a pandas Dataframe whose index is not starting from 0 as the input to
  the ModelVersion.run.

New Features

• Feature Store: Add overloads to APIs accept both object and name/version. Impacted APIs include read_feature_view(),
  refresh_feature_view(), get_refresh_history(), resume_feature_view(), suspend_feature_view(), delete_feature_view().
• Feature Store: Add docstring inline examples for all public APIs.
• Feature Store: Add new utility class ExampleHelper to help with load source data to simplify public notebooks.
• Registry: Option to enable_explainability when registering XGBoost models as a pre-PuPr feature.
• Feature Store: add new API update_entity().
• Registry: Option to enable_explainability when registering Catboost models as a pre-PuPr feature.
• Feature Store: Add new argument warehouse to FeatureView constructor to overwrite the default warehouse. Also add
  a new column 'warehouse' to the output of list_feature_views().
• Registry: Add support for logging model from a model version.
• Modeling: Distributed Hyperparameter Optimization now announce GA refresh version. The latest memory efficient version
  will not have the 10GB training limitation for dataset any more. To turn off, please run
  from snowflake.ml.modeling._internal.snowpark_implementations import ( distributed_hpo_trainer, ) distributed_hpo_trainer.ENABLE_EFFICIENT_MEMORY_USAGE = False
• Registry: Option to enable_explainability when registering LightGBM models as a pre-PuPr feature.

Behavior Changes

• Feature Store: change some positional parameters to keyword arguments in following APIs:
  • Entity(): desc.
  • FeatureView(): timestamp_col, refresh_freq, desc.
  • FeatureStore(): creation_mode.
  • update_entity(): desc.
  • register_feature_view(): block, overwrite.
  • list_feature_views(): entity_name, feature_view_name.
  • get_refresh_history(): verbose.
  • retrieve_feature_values(): spine_timestamp_col, exclude_columns, include_feature_view_timestamp_col.
  • generate_training_set(): save_as, spine_timestamp_col, spine_label_cols, exclude_columns,
    include_feature_view_timestamp_col.
  • generate_dataset(): version, spine_timestamp_col, spine_label_cols, exclude_columns,
    include_feature_view_timestamp_col, desc, output_type.

1.5.4

1.5.4

Bug Fixes

• Model Registry (PrPr): Fix 401 Unauthorized issue when deploying model to SPCS.
• Feature Store: Downgrades exceptions to warnings for few property setters in feature view. Now you can set
  desc, refresh_freq and warehouse for draft feature views.
• Modeling: Fix an issue with calling OrdinalEncoder with categories as a dictionary and a pandas DataFrame
• Modeling: Fix an issue with calling OneHotEncoder with categories as a dictionary and a pandas DataFrame

New Features

• Registry: Allow overriding device_map and device when loading huggingface pipeline models.
• Registry: Add set_alias method to ModelVersion instance to set an alias to model version.
• Registry: Add unset_alias method to ModelVersion instance to unset an alias to model version.
• Registry: Add partitioned_inference_api allowing users to create partitioned inference functions in registered
  models. Enable model inference methods with table functions with vectorized process methods in registered models.
• Feature Store: add 3 more columns: refresh_freq, refresh_mode and scheduling_state to the result of
  list_feature_views().
• Feature Store: update_feature_view() supports updating description.
• Feature Store: add new API refresh_feature_view().
• Feature Store: add new API get_refresh_history().
• Feature Store: Add generate_training_set() API for generating table-backed feature snapshots.
• Feature Store: Add DeprecationWarning for generate_dataset(..., output_type="table").
• Feature Store: update_feature_view() supports updating description.
• Feature Store: add new API refresh_feature_view().
• Feature Store: add new API get_refresh_history().
• Model Development: OrdinalEncoder supports a list of array-likes for categories argument.
• Model Development: OneHotEncoder supports a list of array-likes for categories argument.

1.5.3

Bug Fixes
Modeling: Fix an issue causing lineage information to be missing for Pipeline, GridSearchCV , SimpleImputer, and RandomizedSearchCV
Registry: Fix an issue that leads to incorrect result when using pandas Dataframe with over 100, 000 rows as the input of ModelVersion.run method in Stored Procedure.

New Features
Registry: Add support for TIMESTAMP_NTZ model signature data type, allowing timestamp input and output.
Dataset: Add DatasetVersion.label_cols and DatasetVersion.exclude_cols properties.