FLAML 2.3.1__tar.gz → 2.3.3__tar.gz

{flaml-2.3.1 → flaml-2.3.3}/FLAML.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: FLAML
-Version: 2.3.1
+Version: 2.3.3
 Summary: A fast library for automated machine learning and tuning
 Home-page: https://github.com/microsoft/FLAML
 Author: Microsoft Corporation

{flaml-2.3.1 → flaml-2.3.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: FLAML
-Version: 2.3.1
+Version: 2.3.3
 Summary: A fast library for automated machine learning and tuning
 Home-page: https://github.com/microsoft/FLAML
 Author: Microsoft Corporation

{flaml-2.3.1 → flaml-2.3.3}/flaml/__init__.py

@@ -1,4 +1,5 @@
 import logging
+import warnings
 
 try:
     from flaml.automl import AutoML, logger_formatter
@@ -12,7 +13,8 @@ from flaml.version import __version__
 
 # Set the root logger.
 logger = logging.getLogger(__name__)
-logger.setLevel(logging.INFO)
+if logger.level == logging.NOTSET:
+    logger.setLevel(logging.INFO)
 
 if not has_automl:
-    logger.warning("flaml.automl is not available. Please install flaml[automl] to enable AutoML functionalities.")
+    warnings.warn("flaml.automl is not available. Please install flaml[automl] to enable AutoML functionalities.")

{flaml-2.3.1 → flaml-2.3.3}/flaml/automl/automl.py

@@ -203,7 +203,7 @@ class AutoML(BaseEstimator):
                 * Valid str options depend on different tasks.
                 For classification tasks, valid choices are
                     ["auto", 'stratified', 'uniform', 'time', 'group']. "auto" -> stratified.
-                For regression tasks, valid choices are ["auto", 'uniform', 'time'].
+                For regression tasks, valid choices are ["auto", 'uniform', 'time', 'group'].
                     "auto" -> uniform.
                 For time series forecast tasks, must be "auto" or 'time'.
                 For ranking task, must be "auto" or 'group'.
@@ -739,7 +739,7 @@ class AutoML(BaseEstimator):
                 * Valid str options depend on different tasks.
                 For classification tasks, valid choices are
                     ["auto", 'stratified', 'uniform', 'time', 'group']. "auto" -> stratified.
-                For regression tasks, valid choices are ["auto", 'uniform', 'time'].
+                For regression tasks, valid choices are ["auto", 'uniform', 'time', 'group'].
                     "auto" -> uniform.
                 For time series forecast tasks, must be "auto" or 'time'.
                 For ranking task, must be "auto" or 'group'.
@@ -1358,7 +1358,7 @@ class AutoML(BaseEstimator):
                 * Valid str options depend on different tasks.
                 For classification tasks, valid choices are
                     ["auto", 'stratified', 'uniform', 'time', 'group']. "auto" -> stratified.
-                For regression tasks, valid choices are ["auto", 'uniform', 'time'].
+                For regression tasks, valid choices are ["auto", 'uniform', 'time', 'group'].
                     "auto" -> uniform.
                 For time series forecast tasks, must be "auto" or 'time'.
                 For ranking task, must be "auto" or 'group'.

{flaml-2.3.1 → flaml-2.3.3}/flaml/automl/data.py

@@ -293,7 +293,7 @@ class DataTransformer:
                     y = y.rename(TS_VALUE_COL)
             for column in X.columns:
                 # sklearn\utils\validation.py needs int/float values
-                if X[column].dtype.name in ("object", "category"):
+                if X[column].dtype.name in ("object", "category", "string"):
                     if X[column].nunique() == 1 or X[column].nunique(dropna=True) == n - X[column].isnull().sum():
                         X.drop(columns=column, inplace=True)
                         drop = True

{flaml-2.3.1 → flaml-2.3.3}/flaml/automl/model.py

@@ -157,7 +157,25 @@ class BaseEstimator:
 
     @property
     def estimator(self):
-        """Trained model after fit() is called, or None before fit() is called."""
+        """
+        Get the best trained estimator model.
+
+        Returns:
+            object or None: The trained model obtained after calling the `fit()` method,
+            representing the best estimator found during the training process. If `fit()` has
+            not been called yet, it returns `None`.
+
+        Examples:
+            >>> from flaml import AutoML
+            >>> automl = AutoML()
+            >>> automl.fit(X_train, y_train)
+            >>> best_estimator = automl.model.estimator
+            >>> print(best_estimator)
+            RandomForestClassifier()
+
+            Note:
+            To access the best estimator, use `automl.model.estimator`.
+        """
         return self._model
 
     @property
@@ -249,9 +267,11 @@ class BaseEstimator:
             mem = psutil.virtual_memory() if psutil is not None else None
             try:
                 with limit_resource(
-                    mem.available * (1 - free_mem_ratio) + psutil.Process(os.getpid()).memory_info().rss
-                    if mem is not None
-                    else -1,
+                    (
+                        mem.available * (1 - free_mem_ratio) + psutil.Process(os.getpid()).memory_info().rss
+                        if mem is not None
+                        else -1
+                    ),
                     budget,
                 ):
                     train_time = self._fit(X_train, y_train, **kwargs)
@@ -1534,12 +1554,16 @@ class LGBMEstimator(BaseEstimator):
             if n_iter > 1:
                 max_iter = min(
                     n_iter,
-                    int((budget - time.time() + start_time - self._t1) / self._time_per_iter + 1)
-                    if budget is not None
-                    else n_iter,
-                    int((1 - free_mem_ratio) * mem0 / self._mem_per_iter)
-                    if psutil is not None and self._mem_per_iter > 0
-                    else n_iter,
+                    (
+                        int((budget - time.time() + start_time - self._t1) / self._time_per_iter + 1)
+                        if budget is not None
+                        else n_iter
+                    ),
+                    (
+                        int((1 - free_mem_ratio) * mem0 / self._mem_per_iter)
+                        if psutil is not None and self._mem_per_iter > 0
+                        else n_iter
+                    ),
                 )
                 if trained and max_iter <= self.params[self.ITER_HP]:
                     return time.time() - start_time
@@ -1561,18 +1585,17 @@ class LGBMEstimator(BaseEstimator):
                     callbacks = None
             if callbacks is None:
                 self._fit(X_train, y_train, **kwargs)
-            else:
-                self._fit(X_train, y_train, callbacks=callbacks, **kwargs)
-            if callbacks is None:
                 # for xgboost>=1.6.0, pop callbacks to enable pickle
                 callbacks = self.params.pop("callbacks")
                 self._model.set_params(callbacks=callbacks[:-1])
+            else:
+                self._fit(X_train, y_train, callbacks=callbacks, **kwargs)
             best_iteration = (
                 getattr(self._model.get_booster(), "best_iteration", None)
                 if isinstance(self, XGBoostSklearnEstimator)
                 else self._model.best_iteration_
             )
-            if best_iteration is not None:
+            if best_iteration is not None and best_iteration > 0:
                 self._model.set_params(n_estimators=best_iteration + 1)
         else:
             self._fit(X_train, y_train, **kwargs)
@@ -2043,8 +2066,8 @@ class CatBoostEstimator(BaseEstimator):
             self.estimator_class = CatBoostRegressor
 
     def fit(self, X_train, y_train, budget=None, free_mem_ratio=0, **kwargs):
-        if "is_retrain" in kwargs:
-            kwargs.pop("is_retrain")
+        kwargs.pop("is_retrain", None)
+        kwargs.pop("groups", None)
         start_time = time.time()
         deadline = start_time + budget if budget else np.inf
         train_dir = f"catboost_{str(start_time)}"
@@ -2090,7 +2113,8 @@ class CatBoostEstimator(BaseEstimator):
         if weight is not None:
             kwargs["sample_weight"] = weight
         self._model = model
-        self.params[self.ITER_HP] = self._model.tree_count_
+        # Commented-out line below incorrectly assigned n_estimators - see https://github.com/microsoft/FLAML/pull/1364
+        # self.params[self.ITER_HP] = self._model.tree_count_
         train_time = time.time() - start_time
         return train_time
 
@@ -2184,6 +2208,11 @@ class SVCEstimator(SKLearnEstimator):
 
     def __init__(self, task="binary", **config):
         super().__init__(task, **config)
+        self.params.update(
+            {
+                "random_state": config.get("random_seed", 10242048),
+            }
+        )
         assert self._task.is_classification(), "LinearSVC for classification task only"
         self.estimator_class = LinearSVC
 
@@ -2428,6 +2457,11 @@ class ElasticNetEstimator(SKLearnEstimator):
 
     def __init__(self, task="regression", **config):
         super().__init__(task, **config)
+        self.params.update(
+            {
+                "random_state": config.get("random_seed", 10242048),
+            }
+        )
         assert self._task.is_regression(), "ElasticNet for regression task only"
         self.estimator_class = ElasticNet
 
@@ -2752,7 +2786,7 @@ class BaseResourceLimit:
     def check_resource_limits(self, current_time, current_iteration, mllib):
         if (mllib == "xgb" and current_iteration == 0) or (mllib == "cat" and current_iteration == 1):
             self._time_per_iter = current_time - self.start_time
-        if current_time + self._time_per_iter > self.deadline:
+        if mllib != "cat" and current_time + self._time_per_iter > self.deadline:
             return False
         if psutil is not None and self.free_mem_ratio is not None:
             mem = psutil.virtual_memory()

{flaml-2.3.1 → flaml-2.3.3}/flaml/automl/spark/metrics.py

@@ -1,3 +1,4 @@
+import json
 from typing import Union
 
 import numpy as np
@@ -9,7 +10,7 @@ from pyspark.ml.evaluation import (
     RegressionEvaluator,
 )
 
-from flaml.automl.spark import F, psSeries
+from flaml.automl.spark import F, T, psDataFrame, psSeries, sparkDataFrame
 
 
 def ps_group_counts(groups: Union[psSeries, np.ndarray]) -> np.ndarray:
@@ -36,6 +37,16 @@ def _compute_label_from_probability(df, probability_col, prediction_col):
     return df
 
 
+def string_to_array(s):
+    try:
+        return json.loads(s)
+    except json.JSONDecodeError:
+        return []
+
+
+string_to_array_udf = F.udf(string_to_array, T.ArrayType(T.DoubleType()))
+
+
 def spark_metric_loss_score(
     metric_name: str,
     y_predict: psSeries,
@@ -135,6 +146,11 @@ def spark_metric_loss_score(
         )
     elif metric_name == "log_loss":
         # For log_loss, prediction_col should be probability, and we need to convert it to label
+        # handle data like "{'type': '1', 'values': '[1, 2, 3]'}"
+        # Fix cannot resolve "array_max(prediction)" due to data type mismatch: Parameter 1 requires the "ARRAY" type,
+        # however "prediction" has the type "STRUCT<type: TINYINT, size: INT, indices: ARRAY<INT>, values: ARRAY<DOUBLE>>"
+        df = df.withColumn(prediction_col, df[prediction_col].cast(T.StringType()))
+        df = df.withColumn(prediction_col, string_to_array_udf(df[prediction_col]))
         df = _compute_label_from_probability(df, prediction_col, prediction_col + "_label")
         evaluator = MulticlassClassificationEvaluator(
             metricName="logLoss",

{flaml-2.3.1 → flaml-2.3.3}/flaml/automl/task/generic_task.py

@@ -87,7 +87,6 @@ class GenericTask(Task):
                 "transformer": TransformersEstimator,
                 "transformer_ms": TransformersEstimatorModelSelection,
                 "histgb": HistGradientBoostingEstimator,
-                # Above are open-source, below are internal
                 "svc": SVCEstimator,
                 "sgd": SGDEstimator,
                 "nb_spark": SparkNaiveBayesEstimator,
@@ -443,8 +442,8 @@ class GenericTask(Task):
                 X_train_all, y_train_all = shuffle(X_train_all, y_train_all, random_state=RANDOM_SEED)
             if data_is_df:
                 X_train_all.reset_index(drop=True, inplace=True)
-            if isinstance(y_train_all, pd.Series):
-                y_train_all.reset_index(drop=True, inplace=True)
+        if isinstance(y_train_all, pd.Series):
+            y_train_all.reset_index(drop=True, inplace=True)
 
         X_train, y_train = X_train_all, y_train_all
         state.groups_all = state.groups
@@ -706,7 +705,6 @@ class GenericTask(Task):
             fit_kwargs = {}
         if cv_score_agg_func is None:
             cv_score_agg_func = default_cv_score_agg_func
-        start_time = time.time()
         val_loss_folds = []
         log_metric_folds = []
         metric = None
@@ -813,8 +811,6 @@ class GenericTask(Task):
             if is_spark_dataframe:
                 X_train.spark.unpersist()  # uncache data to free memory
                 X_val.spark.unpersist()  # uncache data to free memory
-            if budget and time.time() - start_time >= budget:
-                break
         val_loss, metric = cv_score_agg_func(val_loss_folds, log_metric_folds)
         n = total_fold_num
         pred_time /= n

{flaml-2.3.1 → flaml-2.3.3}/flaml/automl/task/task.py

@@ -192,7 +192,7 @@ class Task(ABC):
                 * Valid str options depend on different tasks.
                 For classification tasks, valid choices are
                     ["auto", 'stratified', 'uniform', 'time', 'group']. "auto" -> stratified.
-                For regression tasks, valid choices are ["auto", 'uniform', 'time'].
+                For regression tasks, valid choices are ["auto", 'uniform', 'time', 'group'].
                     "auto" -> uniform.
                 For time series forecast tasks, must be "auto" or 'time'.
                 For ranking task, must be "auto" or 'group'.

{flaml-2.3.1 → flaml-2.3.3}/flaml/automl/time_series/ts_data.py

@@ -393,7 +393,7 @@ class DataTransformerTS:
 
         for column in X.columns:
             # sklearn/utils/validation.py needs int/float values
-            if X[column].dtype.name in ("object", "category"):
+            if X[column].dtype.name in ("object", "category", "string"):
                 if (
                     # drop columns where all values are the same
                     X[column].nunique() == 1

{flaml-2.3.1 → flaml-2.3.3}/flaml/fabric/mlflow.py

@@ -127,6 +127,13 @@ def _get_notebook_name():
     return None
 
 
+def safe_json_dumps(obj):
+    def default(o):
+        return str(o)
+
+    return json.dumps(obj, default=default)
+
+
 class MLflowIntegration:
     def __init__(self, experiment_type="automl", mlflow_exp_name=None, extra_tag=None):
         try:
@@ -438,7 +445,7 @@ class MLflowIntegration:
                 "flaml.meric": automl_metric_name,
                 "flaml.run_source": "flaml-automl",
                 "flaml.log_type": self.log_type,
-                "flaml.automl_user_configurations": json.dumps(automl._automl_user_configurations),
+                "flaml.automl_user_configurations": safe_json_dumps(automl._automl_user_configurations),
             },
             "params": {
                 "sample_size": search_state.sample_size,

{flaml-2.3.1 → flaml-2.3.3}/flaml/tune/tune.py

@@ -260,6 +260,8 @@ def run(
     mlflow_exp_name: Optional[str] = None,
     automl_info: Optional[Tuple[float]] = None,
     extra_tag: Optional[dict] = None,
+    cost_attr: Optional[str] = "auto",
+    cost_budget: Optional[float] = None,
     **ray_args,
 ):
     """The function-based way of performing HPO.
@@ -462,6 +464,12 @@ def run(
             overwritten by the value of `n_concurrent_trials` in AutoML. When <= 0, the concurrent trials
             will be set to the number of executors.
         extra_tag: dict, default=None | Extra tags to be added to the mlflow runs created by autologging.
+        cost_attr: None or str to specify the attribute to evaluate the cost of different trials.
+            Default is "auto", which means that we will automatically choose the cost attribute to use (depending
+            on the nature of the resource budget). When cost_attr is set to None, cost differences between different trials will be omitted
+            in our search algorithm. When cost_attr is set to a str different from "auto" and "time_total_s",
+            this cost_attr must be available in the result dict of the trial.
+        cost_budget: A float of the cost budget. Only valid when cost_attr is a str different from "auto" and "time_total_s".
         **ray_args: keyword arguments to pass to ray.tune.run().
             Only valid when use_ray=True.
     """
@@ -600,6 +608,8 @@ def run(
             metric_constraints=metric_constraints,
             use_incumbent_result_in_evaluation=use_incumbent_result_in_evaluation,
             lexico_objectives=lexico_objectives,
+            cost_attr=cost_attr,
+            cost_budget=cost_budget,
         )
     else:
         if metric is None or mode is None:

flaml-2.3.3/flaml/version.py

@@ -0,0 +1 @@
+__version__ = "2.3.3"

flaml-2.3.1/flaml/version.py

@@ -1 +0,0 @@
-__version__ = "2.3.1"