autogluon.tabular 1.5.0b20251228__py3-none-any.whl → 1.5.1b20260116__py3-none-any.whl

autogluon/tabular/__init__.py

@@ -1,5 +1,6 @@
 # noinspection PyUnresolvedReferences
 from autogluon.common.dataset import TabularDataset
+
 # noinspection PyUnresolvedReferences
 from autogluon.common.features.feature_metadata import FeatureMetadata
 from autogluon.common.utils.log_utils import _add_stream_handler

autogluon/tabular/configs/config_helper.py

@@ -125,7 +125,9 @@ class ConfigBuilder:
 
         if isinstance(presets, list):
             unknown_keys = [k for k in presets if k not in valid_keys]
-            assert len(unknown_keys) == 0, f"The following presets are not recognized: {unknown_keys} - use one of the valid presets: {valid_keys}"
+            assert len(unknown_keys) == 0, (
+                f"The following presets are not recognized: {unknown_keys} - use one of the valid presets: {valid_keys}"
+            )
 
         self.config["presets"] = presets
         return self
@@ -144,12 +146,18 @@ class ConfigBuilder:
         valid_keys = self._valid_keys()
         valid_str_values = list(hyperparameter_config_dict.keys())
         if isinstance(hyperparameters, str):
-            assert hyperparameters in hyperparameter_config_dict, f"{hyperparameters} is not one of the valid presets {valid_str_values}"
+            assert hyperparameters in hyperparameter_config_dict, (
+                f"{hyperparameters} is not one of the valid presets {valid_str_values}"
+            )
         elif isinstance(hyperparameters, dict):
             unknown_keys = [k for k in hyperparameters.keys() if isinstance(k, str) and (k not in valid_keys)]
-            assert len(unknown_keys) == 0, f"The following model types are not recognized: {unknown_keys} - use one of the valid models: {valid_keys}"
+            assert len(unknown_keys) == 0, (
+                f"The following model types are not recognized: {unknown_keys} - use one of the valid models: {valid_keys}"
+            )
         else:
-            raise ValueError(f"hyperparameters must be either str: {valid_str_values} or dict with keys of {valid_keys}")
+            raise ValueError(
+                f"hyperparameters must be either str: {valid_str_values} or dict with keys of {valid_keys}"
+            )
         self.config["hyperparameters"] = hyperparameters
         return self
 
@@ -230,7 +238,9 @@ class ConfigBuilder:
         """
         valid_str_values = scheduler_factory._scheduler_presets.keys()
         if isinstance(hyperparameter_tune_kwargs, str):
-            assert hyperparameter_tune_kwargs in valid_str_values, f"{hyperparameter_tune_kwargs} string must be one of {valid_str_values}"
+            assert hyperparameter_tune_kwargs in valid_str_values, (
+                f"{hyperparameter_tune_kwargs} string must be one of {valid_str_values}"
+            )
         elif not isinstance(hyperparameter_tune_kwargs, dict):
             raise ValueError(f"hyperparameter_tune_kwargs must be either str: {valid_str_values} or dict")
         self.config["hyperparameter_tune_kwargs"] = hyperparameter_tune_kwargs
@@ -294,7 +304,9 @@ class ConfigBuilder:
             models = [models]
 
         unknown_keys = [k for k in models if isinstance(k, str) and (k not in valid_keys)]
-        assert len(unknown_keys) == 0, f"The following model types are not recognized: {unknown_keys} - use one of the valid models: {valid_keys}"
+        assert len(unknown_keys) == 0, (
+            f"The following model types are not recognized: {unknown_keys} - use one of the valid models: {valid_keys}"
+        )
 
         models = [m for m in valid_keys if m not in models]
         self.config["excluded_model_types"] = models

autogluon/tabular/configs/feature_generator_presets.py

@@ -18,7 +18,9 @@ def get_default_feature_generator(feature_generator, feature_metadata=None, init
         elif feature_generator == "interpretable":
             feature_generator = AutoMLInterpretablePipelineFeatureGenerator(**init_kwargs)
         else:
-            raise ValueError(f"Unknown feature_generator preset: '{feature_generator}', valid presets: {['auto', 'interpretable']}")
+            raise ValueError(
+                f"Unknown feature_generator preset: '{feature_generator}', valid presets: {['auto', 'interpretable']}"
+            )
     if feature_metadata is not None:
         if feature_generator.feature_metadata_in is None and not feature_generator.is_fit():
             feature_generator.feature_metadata_in = copy.deepcopy(feature_metadata)

autogluon/tabular/configs/hyperparameter_configs.py

@@ -27,12 +27,18 @@ hyperparameter_config_dict = dict(
         "RF": [
             {"criterion": "gini", "ag_args": {"name_suffix": "Gini", "problem_types": ["binary", "multiclass"]}},
             {"criterion": "entropy", "ag_args": {"name_suffix": "Entr", "problem_types": ["binary", "multiclass"]}},
-            {"criterion": "squared_error", "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]}},
+            {
+                "criterion": "squared_error",
+                "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]},
+            },
         ],
         "XT": [
             {"criterion": "gini", "ag_args": {"name_suffix": "Gini", "problem_types": ["binary", "multiclass"]}},
             {"criterion": "entropy", "ag_args": {"name_suffix": "Entr", "problem_types": ["binary", "multiclass"]}},
-            {"criterion": "squared_error", "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]}},
+            {
+                "criterion": "squared_error",
+                "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]},
+            },
         ],
     },
     # Results in smaller models. Generally will make inference speed much faster and disk usage much lower, but with worse accuracy.
@@ -53,14 +59,38 @@ hyperparameter_config_dict = dict(
         "XGB": {},
         "FASTAI": {},
         "RF": [
-            {"criterion": "gini", "max_depth": 15, "ag_args": {"name_suffix": "Gini", "problem_types": ["binary", "multiclass"]}},
-            {"criterion": "entropy", "max_depth": 15, "ag_args": {"name_suffix": "Entr", "problem_types": ["binary", "multiclass"]}},
-            {"criterion": "squared_error", "max_depth": 15, "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]}},
+            {
+                "criterion": "gini",
+                "max_depth": 15,
+                "ag_args": {"name_suffix": "Gini", "problem_types": ["binary", "multiclass"]},
+            },
+            {
+                "criterion": "entropy",
+                "max_depth": 15,
+                "ag_args": {"name_suffix": "Entr", "problem_types": ["binary", "multiclass"]},
+            },
+            {
+                "criterion": "squared_error",
+                "max_depth": 15,
+                "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]},
+            },
         ],
         "XT": [
-            {"criterion": "gini", "max_depth": 15, "ag_args": {"name_suffix": "Gini", "problem_types": ["binary", "multiclass"]}},
-            {"criterion": "entropy", "max_depth": 15, "ag_args": {"name_suffix": "Entr", "problem_types": ["binary", "multiclass"]}},
-            {"criterion": "squared_error", "max_depth": 15, "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]}},
+            {
+                "criterion": "gini",
+                "max_depth": 15,
+                "ag_args": {"name_suffix": "Gini", "problem_types": ["binary", "multiclass"]},
+            },
+            {
+                "criterion": "entropy",
+                "max_depth": 15,
+                "ag_args": {"name_suffix": "Entr", "problem_types": ["binary", "multiclass"]},
+            },
+            {
+                "criterion": "squared_error",
+                "max_depth": 15,
+                "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]},
+            },
         ],
     },
     # Results in much smaller models. Behaves similarly to 'light', but in many cases with over 10x less disk usage and a further reduction in accuracy.
@@ -137,6 +167,7 @@ hyperparameter_config_dict["experimental_2024"] = {"TABPFNMIX": tabpfnmix_defaul
 hyperparameter_config_dict["experimental_2024"].update(hyperparameter_config_dict["zeroshot_2023"])
 hyperparameter_config_dict["experimental"] = hyperparameter_config_dict["experimental_2024"]
 
+
 def get_hyperparameter_config_options():
     return list(hyperparameter_config_dict.keys())
 
@@ -144,5 +175,7 @@ def get_hyperparameter_config_options():
 def get_hyperparameter_config(config_name):
     config_options = get_hyperparameter_config_options()
     if config_name not in config_options:
-        raise ValueError(f"Valid hyperparameter config names are: {config_options}, but '{config_name}' was given instead.")
+        raise ValueError(
+            f"Valid hyperparameter config names are: {config_options}, but '{config_name}' was given instead."
+        )
     return copy.deepcopy(hyperparameter_config_dict[config_name])

autogluon/tabular/configs/presets_configs.py

@@ -9,14 +9,18 @@ tabular_presets_dict = dict(
         "hyperparameters": "zeroshot",
         "time_limit": 3600,
     },
-
     best_quality_v150={
         "auto_stack": True,
         "dynamic_stacking": "auto",
         "num_stack_levels": 0,
         "hyperparameters": "zeroshot_2025_12_18_cpu",
         "time_limit": 3600,
-        "callbacks": [["EarlyStoppingCountCallback", {"patience": [[100, 4], [500, 8], [2500, 15], [10000, 40], [100000, 100], None]}]],
+        "callbacks": [
+            [
+                "EarlyStoppingCountCallback",
+                {"patience": [[100, 4], [500, 8], [2500, 15], [10000, 40], [100000, 100], None]},
+            ]
+        ],
     },
     # High predictive accuracy with fast inference. ~8x faster inference and ~8x lower disk usage than `best_quality`.
     # Recommended for applications that require fast inference speed and/or small model size.
@@ -30,19 +34,22 @@ tabular_presets_dict = dict(
         "set_best_to_refit_full": True,
         "save_bag_folds": False,
     },
-
     high_quality_v150={
         "auto_stack": True,
         "dynamic_stacking": "auto",
         "num_stack_levels": 0,
         "hyperparameters": "zeroshot_2025_12_18_cpu",
         "time_limit": 3600,
-        "callbacks": [["EarlyStoppingCountCallback", {"patience": [[100, 4], [500, 8], [2500, 15], [10000, 40], [100000, 100], None]}]],
+        "callbacks": [
+            [
+                "EarlyStoppingCountCallback",
+                {"patience": [[100, 4], [500, 8], [2500, 15], [10000, 40], [100000, 100], None]},
+            ]
+        ],
         "refit_full": True,
         "set_best_to_refit_full": True,
         "save_bag_folds": False,
     },
-
     # Good predictive accuracy with very fast inference. ~4x faster training, ~8x faster inference and ~8x lower disk usage than `high_quality`.
     # Recommended for applications that require very fast inference speed.
     # Aliases: good
@@ -68,7 +75,13 @@ tabular_presets_dict = dict(
     optimize_for_deployment={"keep_only_best": True, "save_space": True},
     # Disables automated feature generation when text features are detected.
     # This is useful to determine how beneficial text features are to the end result, as well as to ensure features are not mistaken for text when they are not.
-    ignore_text={"_feature_generator_kwargs": {"enable_text_ngram_features": False, "enable_text_special_features": False, "enable_raw_text_features": False}},
+    ignore_text={
+        "_feature_generator_kwargs": {
+            "enable_text_ngram_features": False,
+            "enable_text_special_features": False,
+            "enable_raw_text_features": False,
+        }
+    },
     ignore_text_ngrams={"_feature_generator_kwargs": {"enable_text_ngram_features": False}},
     # Fit only interpretable models.
     interpretable={
@@ -86,14 +99,24 @@ tabular_presets_dict = dict(
     best_quality_v082={"auto_stack": True},
     # High predictive accuracy with fast inference. ~10x-200x faster inference and ~10x-200x lower disk usage than `best_quality`.
     # Recommended for applications that require reasonable inference speed and/or model size.
-    high_quality_v082={"auto_stack": True, "refit_full": True, "set_best_to_refit_full": True, "save_bag_folds": False},
+    high_quality_v082={
+        "auto_stack": True,
+        "refit_full": True,
+        "set_best_to_refit_full": True,
+        "save_bag_folds": False,
+    },
     # Good predictive accuracy with very fast inference. ~4x faster inference and ~4x lower disk usage than `high_quality`.
     # Recommended for applications that require fast inference speed.
-    good_quality_v082={"auto_stack": True, "refit_full": True, "set_best_to_refit_full": True, "save_bag_folds": False, "hyperparameters": "light"},
+    good_quality_v082={
+        "auto_stack": True,
+        "refit_full": True,
+        "set_best_to_refit_full": True,
+        "save_bag_folds": False,
+        "hyperparameters": "light",
+    },
     # ------------------------------------------
     # Experimental presets. Only use these presets if you are ok with unstable and potentially poor performing presets.
     #  Experimental presets can be removed or changed without warning.
-
     # [EXPERIMENTAL PRESET] The `extreme` preset may be changed or removed without warning.
     # This preset acts as a testing ground for cutting edge features and models which could later be added to the `best_quality` preset in future releases.
     # Using this preset can lead to unexpected crashes, as it hasn't been as thoroughly tested as other presets.
@@ -110,9 +133,13 @@ tabular_presets_dict = dict(
         "num_stack_levels": 0,
         "hyperparameters": "zeroshot_2025_12_18_gpu",
         "time_limit": 3600,
-        "callbacks": [["EarlyStoppingCountCallback", {"patience": [[100, 4], [500, 8], [2500, 15], [10000, 40], [100000, 100], None]}]],
+        "callbacks": [
+            [
+                "EarlyStoppingCountCallback",
+                {"patience": [[100, 4], [500, 8], [2500, 15], [10000, 40], [100000, 100], None]},
+            ]
+        ],
     },
-
     extreme_quality_v140={
         "auto_stack": True,
         "dynamic_stacking": "auto",
@@ -121,7 +148,6 @@ tabular_presets_dict = dict(
         "hyperparameters": None,
         "time_limit": 3600,
     },
-
     # Preset with a portfolio learned from TabArena v0.1: https://tabarena.ai/
     # Uses tabular foundation models: TabPFNv2, TabICL, Mitra
     # Uses deep learning model: TabM
@@ -136,7 +162,6 @@ tabular_presets_dict = dict(
         "hyperparameters": "zeroshot_2025_tabfm",
         "time_limit": 3600,
     },
-
     # DOES NOT SUPPORT GPU.
     experimental_quality_v120={
         "auto_stack": True,
@@ -147,7 +172,6 @@ tabular_presets_dict = dict(
         "num_gpus": 0,
         "time_limit": 3600,
     },
-
     # ------------------------------------------
     # ------------------------------------------
     # ------------------------------------------

autogluon/tabular/configs/zeroshot/zeroshot_portfolio_2023.py

@@ -773,24 +773,94 @@ hyperparameter_portfolio_zeroshot_2023 = {
         {"criterion": "gini", "ag_args": {"name_suffix": "Gini", "problem_types": ["binary", "multiclass"]}},
         {"criterion": "entropy", "ag_args": {"name_suffix": "Entr", "problem_types": ["binary", "multiclass"]}},
         {"criterion": "squared_error", "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]}},
-        {"max_features": 0.75, "max_leaf_nodes": 37308, "min_samples_leaf": 1, "ag_args": {"name_suffix": "_r195", "priority": -13}},
-        {"max_features": 0.75, "max_leaf_nodes": 28310, "min_samples_leaf": 2, "ag_args": {"name_suffix": "_r39", "priority": -32}},
-        {"max_features": 1.0, "max_leaf_nodes": 38572, "min_samples_leaf": 5, "ag_args": {"name_suffix": "_r127", "priority": -45}},
-        {"max_features": 0.75, "max_leaf_nodes": 18242, "min_samples_leaf": 40, "ag_args": {"name_suffix": "_r34", "priority": -47}},
-        {"max_features": "log2", "max_leaf_nodes": 42644, "min_samples_leaf": 1, "ag_args": {"name_suffix": "_r166", "priority": -63}},
-        {"max_features": 0.75, "max_leaf_nodes": 36230, "min_samples_leaf": 3, "ag_args": {"name_suffix": "_r15", "priority": -68}},
-        {"max_features": 1.0, "max_leaf_nodes": 48136, "min_samples_leaf": 1, "ag_args": {"name_suffix": "_r16", "priority": -81}},
+        {
+            "max_features": 0.75,
+            "max_leaf_nodes": 37308,
+            "min_samples_leaf": 1,
+            "ag_args": {"name_suffix": "_r195", "priority": -13},
+        },
+        {
+            "max_features": 0.75,
+            "max_leaf_nodes": 28310,
+            "min_samples_leaf": 2,
+            "ag_args": {"name_suffix": "_r39", "priority": -32},
+        },
+        {
+            "max_features": 1.0,
+            "max_leaf_nodes": 38572,
+            "min_samples_leaf": 5,
+            "ag_args": {"name_suffix": "_r127", "priority": -45},
+        },
+        {
+            "max_features": 0.75,
+            "max_leaf_nodes": 18242,
+            "min_samples_leaf": 40,
+            "ag_args": {"name_suffix": "_r34", "priority": -47},
+        },
+        {
+            "max_features": "log2",
+            "max_leaf_nodes": 42644,
+            "min_samples_leaf": 1,
+            "ag_args": {"name_suffix": "_r166", "priority": -63},
+        },
+        {
+            "max_features": 0.75,
+            "max_leaf_nodes": 36230,
+            "min_samples_leaf": 3,
+            "ag_args": {"name_suffix": "_r15", "priority": -68},
+        },
+        {
+            "max_features": 1.0,
+            "max_leaf_nodes": 48136,
+            "min_samples_leaf": 1,
+            "ag_args": {"name_suffix": "_r16", "priority": -81},
+        },
     ],
     "XT": [
         {"criterion": "gini", "ag_args": {"name_suffix": "Gini", "problem_types": ["binary", "multiclass"]}},
         {"criterion": "entropy", "ag_args": {"name_suffix": "Entr", "problem_types": ["binary", "multiclass"]}},
         {"criterion": "squared_error", "ag_args": {"name_suffix": "MSE", "problem_types": ["regression", "quantile"]}},
-        {"max_features": 0.75, "max_leaf_nodes": 18392, "min_samples_leaf": 1, "ag_args": {"name_suffix": "_r42", "priority": -9}},
-        {"max_features": 1.0, "max_leaf_nodes": 12845, "min_samples_leaf": 4, "ag_args": {"name_suffix": "_r172", "priority": -23}},
-        {"max_features": "sqrt", "max_leaf_nodes": 28532, "min_samples_leaf": 1, "ag_args": {"name_suffix": "_r49", "priority": -43}},
-        {"max_features": 1.0, "max_leaf_nodes": 19935, "min_samples_leaf": 20, "ag_args": {"name_suffix": "_r4", "priority": -53}},
-        {"max_features": 0.75, "max_leaf_nodes": 29813, "min_samples_leaf": 4, "ag_args": {"name_suffix": "_r178", "priority": -62}},
-        {"max_features": 1.0, "max_leaf_nodes": 40459, "min_samples_leaf": 1, "ag_args": {"name_suffix": "_r197", "priority": -78}},
-        {"max_features": "sqrt", "max_leaf_nodes": 29702, "min_samples_leaf": 2, "ag_args": {"name_suffix": "_r126", "priority": -86}},
+        {
+            "max_features": 0.75,
+            "max_leaf_nodes": 18392,
+            "min_samples_leaf": 1,
+            "ag_args": {"name_suffix": "_r42", "priority": -9},
+        },
+        {
+            "max_features": 1.0,
+            "max_leaf_nodes": 12845,
+            "min_samples_leaf": 4,
+            "ag_args": {"name_suffix": "_r172", "priority": -23},
+        },
+        {
+            "max_features": "sqrt",
+            "max_leaf_nodes": 28532,
+            "min_samples_leaf": 1,
+            "ag_args": {"name_suffix": "_r49", "priority": -43},
+        },
+        {
+            "max_features": 1.0,
+            "max_leaf_nodes": 19935,
+            "min_samples_leaf": 20,
+            "ag_args": {"name_suffix": "_r4", "priority": -53},
+        },
+        {
+            "max_features": 0.75,
+            "max_leaf_nodes": 29813,
+            "min_samples_leaf": 4,
+            "ag_args": {"name_suffix": "_r178", "priority": -62},
+        },
+        {
+            "max_features": 1.0,
+            "max_leaf_nodes": 40459,
+            "min_samples_leaf": 1,
+            "ag_args": {"name_suffix": "_r197", "priority": -78},
+        },
+        {
+            "max_features": "sqrt",
+            "max_leaf_nodes": 29702,
+            "min_samples_leaf": 2,
+            "ag_args": {"name_suffix": "_r126", "priority": -86},
+        },
     ],
 }