autogluon.tabular 1.3.2b20250722__py3-none-any.whl → 1.4.0b20250724__py3-none-any.whl

autogluon/tabular/models/mitra/sklearn_interface.py

@@ -1,3 +1,6 @@
+from __future__ import annotations
+
+import os
 import time
 from pathlib import Path
 import contextlib
@@ -76,6 +79,7 @@ class MitraBase(BaseEstimator):
             random_mirror_regression=RANDOM_MIRROR_REGRESSION,
             random_mirror_x=RANDOM_MIRROR_X,
             seed=SEED,
+            verbose=True,
         ):
         """
         Initialize the base Mitra model.
@@ -114,8 +118,11 @@ class MitraBase(BaseEstimator):
         self.trainers = []
         self.train_time = 0
         self.seed = seed
+        self.verbose = verbose
 
-        set_seed(self.seed)
+        # FIXME: set_seed was removed in v1.4 as quality and speed reduction was observed when setting seed.
+        #  This should be investigated and fixed for v1.5
+        # set_seed(self.seed)
 
     def _create_config(self, task, dim_output, time_limit=None):
         cfg = ConfigRun(
@@ -183,6 +190,7 @@ class MitraBase(BaseEstimator):
         """Train the ensemble of models."""
 
         cfg, Tab2D = self._create_config(task, dim_output, time_limit)
+        rng = np.random.RandomState(cfg.seed)
 
         success = False
         while not (success and cfg.hyperparams["max_samples_support"] > 0 and cfg.hyperparams["max_samples_query"] > 0):
@@ -217,7 +225,7 @@ class MitraBase(BaseEstimator):
                             path_to_weights=Path(self.state_dict),
                             device=self.device,
                         )
-                    trainer = TrainerFinetune(cfg, model, n_classes=n_classes, device=self.device)
+                    trainer = TrainerFinetune(cfg, model, n_classes=n_classes, device=self.device, rng=rng, verbose=self.verbose)
 
                     start_time = time.time()
                     trainer.train(X_train, y_train, X_valid, y_valid)
@@ -275,6 +283,7 @@ class MitraClassifier(MitraBase, ClassifierMixin):
             random_mirror_regression=RANDOM_MIRROR_REGRESSION,
             random_mirror_x=RANDOM_MIRROR_X,
             seed=SEED,
+            verbose=True,
         ):
         """Initialize the classifier."""
         super().__init__(
@@ -294,6 +303,7 @@ class MitraClassifier(MitraBase, ClassifierMixin):
             random_mirror_regression=random_mirror_regression,
             random_mirror_x=random_mirror_x,
             seed=seed,
+            verbose=verbose,
         )
         self.task = 'classification'
 
@@ -403,6 +413,7 @@ class MitraRegressor(MitraBase, RegressorMixin):
             random_mirror_regression=RANDOM_MIRROR_REGRESSION,
             random_mirror_x=RANDOM_MIRROR_X,
             seed=SEED,
+            verbose=True,
         ):
         """Initialize the regressor."""
         super().__init__(
@@ -422,6 +433,7 @@ class MitraRegressor(MitraBase, RegressorMixin):
             random_mirror_regression=random_mirror_regression,
             random_mirror_x=random_mirror_x,
             seed=seed,
+            verbose=verbose,
         )
         self.task = 'regression'
 
@@ -492,14 +504,4 @@ class MitraRegressor(MitraBase, RegressorMixin):
 @contextlib.contextmanager
 def mitra_deterministic_context():
     """Context manager to set deterministic settings only for Mitra operations."""
-    
-    original_deterministic_algorithms_set = False
-
-    try:
-        torch.use_deterministic_algorithms(True)
-        original_deterministic_algorithms_set = True
-        yield
-        
-    finally:
-        if original_deterministic_algorithms_set:
-            torch.use_deterministic_algorithms(False)
+    yield

autogluon/tabular/models/tabicl/tabicl_model.py

@@ -98,8 +98,8 @@ class TabICLModel(AbstractModel):
         default_auxiliary_params = super()._get_default_auxiliary_params()
         default_auxiliary_params.update(
             {
-                "max_rows": 100000,
-                "max_features": 500,
+                "max_rows": 30000,
+                "max_features": 2000,
             }
         )
         return default_auxiliary_params
@@ -147,7 +147,7 @@ class TabICLModel(AbstractModel):
         model_mem_estimate *= 1.3  # add 30% buffer
 
         # TODO: Observed memory spikes above expected values on large datasets, increasing mem estimate to compensate
-        model_mem_estimate *= 1.5
+        model_mem_estimate *= 2.0  # Note: 1.5 is not large enough, still gets OOM
 
         mem_estimate = model_mem_estimate + dataset_size_mem_est + baseline_overhead_mem_est
 

autogluon/tabular/models/tabm/rtdl_num_embeddings.py

@@ -1,6 +1,8 @@
 # taken from https://github.com/yandex-research/rtdl-num-embeddings/blob/main/package/rtdl_num_embeddings.py
 """On Embeddings for Numerical Features in Tabular Deep Learning."""
 
+from __future__ import annotations
+
 __version__ = '0.0.12'
 
 __all__ = [
@@ -12,6 +14,7 @@ __all__ = [
     'compute_bins',
 ]
 
+
 import math
 import warnings
 from typing import Any, Literal, Optional, Union

autogluon/tabular/models/tabm/tabm_reference.py

@@ -3,6 +3,8 @@
 # NOTE
 # The minimum required versions of the dependencies are specified in README.md.
 
+from __future__ import annotations
+
 import itertools
 from typing import Any, Literal, Union
 

autogluon/tabular/models/tabpfnv2/tabpfnv2_model.py

@@ -119,12 +119,14 @@ class TabPFNV2Model(AbstractModel):
         super().__init__(**kwargs)
         self._feature_generator = None
         self._cat_features = None
+        self._cat_indices = None
 
     def _preprocess(self, X: pd.DataFrame, is_train=False, **kwargs) -> pd.DataFrame:
         X = super()._preprocess(X, **kwargs)
-        self._cat_indices = []
 
         if is_train:
+            self._cat_indices = []
+
             # X will be the training data.
             self._feature_generator = LabelEncoderFeatureGenerator(verbosity=0)
             self._feature_generator.fit(X=X)
@@ -136,10 +138,11 @@ class TabPFNV2Model(AbstractModel):
                 X=X
             )
 
-            # Detect/set cat features and indices
-            if self._cat_features is None:
-                self._cat_features = self._feature_generator.features_in[:]
-            self._cat_indices = [X.columns.get_loc(col) for col in self._cat_features]
+            if is_train:
+                # Detect/set cat features and indices
+                if self._cat_features is None:
+                    self._cat_features = self._feature_generator.features_in[:]
+                self._cat_indices = [X.columns.get_loc(col) for col in self._cat_features]
 
         return X
 
@@ -187,6 +190,12 @@ class TabPFNV2Model(AbstractModel):
             # logs "Built with PriorLabs-TabPFN"
             self._log_license(device=device)
 
+        if num_gpus == 0:
+            logger.log(
+                30,
+                f"\tWARNING: Running TabPFNv2 on CPU. This can be very slow. We recommend using a GPU instead."
+            )
+
         X = self.preprocess(X, is_train=True)
 
         hps = self._get_model_params()
@@ -366,7 +375,7 @@ class TabPFNV2Model(AbstractModel):
 
         # Add some buffer to each term + 1 GB overhead to be safe
         return int(
-            model_mem + 4 * X_mem + 1.5 * activation_mem + baseline_overhead_mem_est
+            model_mem + 4 * X_mem + 2 * activation_mem + baseline_overhead_mem_est
         )
 
     @classmethod

autogluon/tabular/predictor/predictor.py

@@ -1068,7 +1068,7 @@ class TabularPredictor:
                 20,
                 "No presets specified! To achieve strong results with AutoGluon, it is recommended to use the available presets. Defaulting to `'medium'`...\n"
                 "\tRecommended Presets (For more details refer to https://auto.gluon.ai/stable/tutorials/tabular/tabular-essentials.html#presets):\n"
-                "\tpresets='experimental' : New in v1.2: Pre-trained foundation model + parallel fits. The absolute best accuracy without consideration for inference speed. Does not support GPU.\n"
+                "\tpresets='experimental' : New in v1.4: Massively better than 'best' on datasets <10000 samples by using new models: TabPFNv2, TabICL, Mitra, and TabM. Absolute best accuracy. Requires a GPU. Recommended 64 GB CPU memory and 32+ GB GPU memory.\n"
                 "\tpresets='best'         : Maximize accuracy. Recommended for most users. Use in competitions and benchmarks.\n"
                 "\tpresets='high'         : Strong accuracy with fast inference speed.\n"
                 "\tpresets='good'         : Good accuracy with very fast inference speed.\n"
@@ -1127,10 +1127,48 @@ class TabularPredictor:
         )
         infer_limit, infer_limit_batch_size = self._validate_infer_limit(infer_limit=infer_limit, infer_limit_batch_size=infer_limit_batch_size)
 
+        # TODO: Temporary for v1.4. Make this more extensible for v1.5 by letting users make their own dynamic hyperparameters.
+        dynamic_hyperparameters = kwargs["_experimental_dynamic_hyperparameters"]
+        if dynamic_hyperparameters:
+            logger.log(20, f"Experimental preset uses a dynamic portfolio based on dataset size...")
+            assert hyperparameters is None, f"hyperparameters must be unspecified when `_experimental_dynamic_hyperparameters=True`."
+            n_samples = len(train_data)
+            if n_samples > 30000:
+                data_size = "large"
+            else:
+                data_size = "small"
+            assert data_size in ["large", "small"]
+            if data_size == "large":
+                logger.log(20, f"\tDetected data size: large (>30000 samples), using `zeroshot` portfolio (identical to 'best_quality' preset).")
+                hyperparameters = "zeroshot"
+            else:
+                if "num_stack_levels" not in kwargs_orig:
+                    # disable stacking for tabfm portfolio
+                    num_stack_levels = 0
+                    kwargs["num_stack_levels"] = 0
+                logger.log(
+                    20,
+                    f"\tDetected data size: small (<=30000 samples), using `zeroshot_2025_tabfm` portfolio."
+                    f"\n\t\tNote: `zeroshot_2025_tabfm` portfolio requires a CUDA compatible GPU for best performance."
+                    f"\n\t\tMake sure you have all the relevant dependencies installed: "
+                    f"`pip install autogluon.tabular[tabarena]`."
+                    f"\n\t\tIt is strongly recommended to use a machine with 64+ GB memory "
+                    f"and a CUDA compatible GPU with 32+ GB vRAM when using this preset. "
+                    f"\n\t\tThis portfolio will download foundation model weights from HuggingFace during training. "
+                    f"Ensure you have an internet connection or have pre-downloaded the weights to use these models."
+                    f"\n\t\tThis portfolio was meta-learned with TabArena: https://tabarena.ai"
+                )
+                hyperparameters = "zeroshot_2025_tabfm"
+
         if hyperparameters is None:
             hyperparameters = "default"
         if isinstance(hyperparameters, str):
+            hyperparameters_str = hyperparameters
             hyperparameters = get_hyperparameter_config(hyperparameters)
+            logger.log(
+                20,
+                f"Using hyperparameters preset: hyperparameters='{hyperparameters_str}'",
+            )
         self._validate_hyperparameters(hyperparameters=hyperparameters)
         self.fit_hyperparameters_ = hyperparameters
 
@@ -5042,6 +5080,8 @@ class TabularPredictor:
             learning_curves=False,
             test_data=None,
             raise_on_model_failure=False,
+            # experimental
+            _experimental_dynamic_hyperparameters=False,
         )
         kwargs, ds_valid_keys = self._sanitize_dynamic_stacking_kwargs(kwargs)
         kwargs = self._validate_fit_extra_kwargs(kwargs, extra_valid_keys=list(fit_kwargs_default.keys()) + ds_valid_keys)

autogluon/tabular/trainer/abstract_trainer.py

@@ -2131,6 +2131,8 @@ class AbstractTabularTrainer(AbstractTrainer[AbstractModel]):
         if isinstance(model, BaggedEnsembleModel) and not compute_score:
             # Do not perform OOF predictions when we don't compute a score.
             model_fit_kwargs["_skip_oof"] = True
+        if not isinstance(model, BaggedEnsembleModel):
+            model_fit_kwargs.setdefault("log_resources", True)
 
         model_fit_kwargs = dict(
             model=model,

autogluon/tabular/version.py

@@ -1,4 +1,4 @@
 """This is the autogluon version file."""
 
-__version__ = "1.3.2b20250722"
+__version__ = "1.4.0b20250724"
 __lite__ = False

{autogluon.tabular-1.3.2b20250722.dist-info → autogluon.tabular-1.4.0b20250724.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: autogluon.tabular
-Version: 1.3.2b20250722
+Version: 1.4.0b20250724
 Summary: Fast and Accurate ML in 3 Lines of Code
 Home-page: https://github.com/autogluon/autogluon
 Author: AutoGluon Community
@@ -41,20 +41,22 @@ Requires-Dist: scipy<1.17,>=1.5.4
 Requires-Dist: pandas<2.4.0,>=2.0.0
 Requires-Dist: scikit-learn<1.8.0,>=1.4.0
 Requires-Dist: networkx<4,>=3.0
-Requires-Dist: autogluon.core==1.3.2b20250722
-Requires-Dist: autogluon.features==1.3.2b20250722
+Requires-Dist: autogluon.core==1.4.0b20250724
+Requires-Dist: autogluon.features==1.4.0b20250724
 Provides-Extra: all
-Requires-Dist: einops<0.9,>=0.7; extra == "all"
+Requires-Dist: transformers; extra == "all"
 Requires-Dist: huggingface-hub[torch]; extra == "all"
-Requires-Dist: spacy<3.9; extra == "all"
 Requires-Dist: numpy<2.3.0,>=1.25; extra == "all"
-Requires-Dist: catboost<1.3,>=1.2; extra == "all"
-Requires-Dist: pytabkit<1.6,>=1.5; extra == "all"
-Requires-Dist: xgboost<3.1,>=2.0; extra == "all"
-Requires-Dist: autogluon.core[all]==1.3.2b20250722; extra == "all"
+Requires-Dist: autogluon.core[all]==1.4.0b20250724; extra == "all"
+Requires-Dist: spacy<3.9; extra == "all"
 Requires-Dist: fastai<2.9,>=2.3.1; extra == "all"
 Requires-Dist: torch<2.8,>=2.2; extra == "all"
+Requires-Dist: catboost<1.3,>=1.2; extra == "all"
+Requires-Dist: xgboost<3.1,>=2.0; extra == "all"
+Requires-Dist: loguru; extra == "all"
+Requires-Dist: einx; extra == "all"
 Requires-Dist: lightgbm<4.7,>=4.0; extra == "all"
+Requires-Dist: omegaconf; extra == "all"
 Requires-Dist: blis<1.2.1,>=0.7.0; (platform_system == "Windows" and python_version == "3.9") and extra == "all"
 Provides-Extra: catboost
 Requires-Dist: numpy<2.3.0,>=1.25; extra == "catboost"
@@ -72,9 +74,11 @@ Provides-Extra: mitra
 Requires-Dist: loguru; extra == "mitra"
 Requires-Dist: einx; extra == "mitra"
 Requires-Dist: omegaconf; extra == "mitra"
+Requires-Dist: torch<2.8,>=2.2; extra == "mitra"
 Requires-Dist: transformers; extra == "mitra"
+Requires-Dist: huggingface-hub[torch]; extra == "mitra"
 Provides-Extra: ray
-Requires-Dist: autogluon.core[all]==1.3.2b20250722; extra == "ray"
+Requires-Dist: autogluon.core[all]==1.4.0b20250724; extra == "ray"
 Provides-Extra: realmlp
 Requires-Dist: pytabkit<1.6,>=1.5; extra == "realmlp"
 Provides-Extra: skex
@@ -85,6 +89,24 @@ Requires-Dist: onnxruntime<1.20.0,>=1.17.0; extra == "skl2onnx"
 Requires-Dist: onnxruntime-gpu<1.20.0,>=1.17.0; extra == "skl2onnx"
 Requires-Dist: onnx<1.18.0,>=1.13.0; platform_system != "Windows" and extra == "skl2onnx"
 Requires-Dist: onnx<1.16.2,>=1.13.0; platform_system == "Windows" and extra == "skl2onnx"
+Provides-Extra: tabarena
+Requires-Dist: tabpfn<2.2,>=2.0.9; extra == "tabarena"
+Requires-Dist: transformers; extra == "tabarena"
+Requires-Dist: huggingface-hub[torch]; extra == "tabarena"
+Requires-Dist: pytabkit<1.6,>=1.5; extra == "tabarena"
+Requires-Dist: tabicl<0.2,>=0.1.3; extra == "tabarena"
+Requires-Dist: numpy<2.3.0,>=1.25; extra == "tabarena"
+Requires-Dist: autogluon.core[all]==1.4.0b20250724; extra == "tabarena"
+Requires-Dist: spacy<3.9; extra == "tabarena"
+Requires-Dist: fastai<2.9,>=2.3.1; extra == "tabarena"
+Requires-Dist: torch<2.8,>=2.2; extra == "tabarena"
+Requires-Dist: catboost<1.3,>=1.2; extra == "tabarena"
+Requires-Dist: xgboost<3.1,>=2.0; extra == "tabarena"
+Requires-Dist: loguru; extra == "tabarena"
+Requires-Dist: einx; extra == "tabarena"
+Requires-Dist: lightgbm<4.7,>=4.0; extra == "tabarena"
+Requires-Dist: omegaconf; extra == "tabarena"
+Requires-Dist: blis<1.2.1,>=0.7.0; (platform_system == "Windows" and python_version == "3.9") and extra == "tabarena"
 Provides-Extra: tabicl
 Requires-Dist: tabicl<0.2,>=0.1.3; extra == "tabicl"
 Provides-Extra: tabm
@@ -96,16 +118,16 @@ Requires-Dist: torch<2.8,>=2.2; extra == "tabpfnmix"
 Requires-Dist: huggingface-hub[torch]; extra == "tabpfnmix"
 Requires-Dist: einops<0.9,>=0.7; extra == "tabpfnmix"
 Provides-Extra: tests
+Requires-Dist: tabicl<0.2,>=0.1.3; extra == "tests"
 Requires-Dist: tabpfn<2.2,>=2.0.9; extra == "tests"
+Requires-Dist: pytabkit<1.6,>=1.5; extra == "tests"
+Requires-Dist: torch<2.8,>=2.2; extra == "tests"
+Requires-Dist: huggingface-hub[torch]; extra == "tests"
+Requires-Dist: einops<0.9,>=0.7; extra == "tests"
 Requires-Dist: imodels<2.1.0,>=1.3.10; extra == "tests"
 Requires-Dist: skl2onnx<1.18.0,>=1.15.0; extra == "tests"
 Requires-Dist: onnxruntime<1.20.0,>=1.17.0; extra == "tests"
 Requires-Dist: onnxruntime-gpu<1.20.0,>=1.17.0; extra == "tests"
-Requires-Dist: tabicl<0.2,>=0.1.3; extra == "tests"
-Requires-Dist: loguru; extra == "tests"
-Requires-Dist: einx; extra == "tests"
-Requires-Dist: omegaconf; extra == "tests"
-Requires-Dist: transformers; extra == "tests"
 Requires-Dist: onnx<1.18.0,>=1.13.0; platform_system != "Windows" and extra == "tests"
 Requires-Dist: onnx<1.16.2,>=1.13.0; platform_system == "Windows" and extra == "tests"
 Provides-Extra: xgboost

{autogluon.tabular-1.3.2b20250722.dist-info → autogluon.tabular-1.4.0b20250724.dist-info}/RECORD

@@ -1,13 +1,14 @@
-autogluon.tabular-1.3.2b20250722-py3.9-nspkg.pth,sha256=cQGwpuGPqg1GXscIwt-7PmME1OnSpD-7ixkikJ31WAY,554
+autogluon.tabular-1.4.0b20250724-py3.9-nspkg.pth,sha256=cQGwpuGPqg1GXscIwt-7PmME1OnSpD-7ixkikJ31WAY,554
 autogluon/tabular/__init__.py,sha256=2OXpJCvENRHubBTYNIPpHX93WWuFZzsJBtTZbNVHVas,400
-autogluon/tabular/version.py,sha256=CdIaQfJPPjnE9mEFR_3W2YJdiCuNqvKW-cgu00I6i3w,91
+autogluon/tabular/version.py,sha256=sQbAMP9TVXNv7nE2ealK-zzy0cZUUxkxcLTpbykHhxU,91
 autogluon/tabular/configs/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/configs/config_helper.py,sha256=JsdVGmpcYL88GPKBznPtqJ1sGaByOSvLn7KWU-HyVoQ,21085
 autogluon/tabular/configs/feature_generator_presets.py,sha256=EV5Ym8VW15q92MwOUpTi7wZFS2QooM51fLg3RdUsn-M,1223
-autogluon/tabular/configs/hyperparameter_configs.py,sha256=hp8J7g5GY3Couz929f1ItawobCw-isLTZJBcLoJY348,18035
-autogluon/tabular/configs/presets_configs.py,sha256=2Jlq1X9sVmVlyUxWsZpDV7ma2TncH5Y2HXDML7x2gYc,6810
+autogluon/tabular/configs/hyperparameter_configs.py,sha256=aQ1rrF8P0MX4Ic5M33O96JtKV-K7YpDrgJmWhYmEyug,6848
+autogluon/tabular/configs/presets_configs.py,sha256=hIT9CgdIol5y-YLKrcupXWySP_P9HZp_ky_vXmFrFWk,7682
 autogluon/tabular/configs/zeroshot/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autogluon/tabular/configs/zeroshot/zeroshot_portfolio_2023.py,sha256=oKO_2nEpI_EiLaUGmNN-3kPBIp5ATndbCOaVZ1m0048,29911
+autogluon/tabular/configs/zeroshot/zeroshot_portfolio_2023.py,sha256=6yd84vPqOk-6sLCoM_e_PlphrR2NZUjliS7L1SMKMug,29777
+autogluon/tabular/configs/zeroshot/zeroshot_portfolio_2025.py,sha256=1EH54KvJT35xNVegh1SuvBWt0Qx07vQUdHyc10TwaOI,11966
 autogluon/tabular/experimental/__init__.py,sha256=PpkdMSv_pPZted1XRIuzcFWKjM-66VMUukTnCcoiW0s,100
 autogluon/tabular/experimental/_scikit_mixin.py,sha256=cKeCmtURAXZnhQGrkCBw5rmACCQF7biAWTT3qX8bM2Q,2281
 autogluon/tabular/experimental/_tabular_classifier.py,sha256=7lGoFdvkHiZS3VpcXo97q4ENV9qyIVDExlWkm0wzL3s,2527
@@ -68,8 +69,8 @@ autogluon/tabular/models/lr/hyperparameters/__init__.py,sha256=47DEQpj8HBSa-_TIm
 autogluon/tabular/models/lr/hyperparameters/parameters.py,sha256=Hr5YC13zjbt3CfCbzGj8iXUIuDn-Q7FvDT2uSuiSVlM,1414
 autogluon/tabular/models/lr/hyperparameters/searchspaces.py,sha256=Igywc-B6qJ9EBLdasrDhW-Ot5FGirIzbXLwv5HRe5Xo,276
 autogluon/tabular/models/mitra/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autogluon/tabular/models/mitra/mitra_model.py,sha256=SQehZMs9WBP9o51CJ6L8njuEq0UcunXbbOM3eedhokw,10263
-autogluon/tabular/models/mitra/sklearn_interface.py,sha256=p3keXfA3cdQhoT24jAytrC1uHOK8tQXpE2tSWagAQzU,18588
+autogluon/tabular/models/mitra/mitra_model.py,sha256=ha-54y_LgEysUpcPr7dB76fETu9kEQ2oLUXteGYVVeE,11927
+autogluon/tabular/models/mitra/sklearn_interface.py,sha256=Znwx1uMagauu1DwcutM_kgGY8maQrxOE0KsP1uS46qE,18751
 autogluon/tabular/models/mitra/_internal/__init__.py,sha256=dN2dz1pGMgQTFiSf9oYbyq23iJUxV8QNlOX3qw3KUO4,35
 autogluon/tabular/models/mitra/_internal/config/__init__.py,sha256=Exu_Sx6-K-D5peDQ_TibsjZpqAALs2-9IXfq8hu1mwU,40
 autogluon/tabular/models/mitra/_internal/config/config_pretrain.py,sha256=CeaD96EcDX69LdcLTYGlFmYLdBNINEJXRMWmJ6LbhTg,6038
@@ -81,16 +82,16 @@ autogluon/tabular/models/mitra/_internal/core/get_loss.py,sha256=hv0t7zvyZ-DgA5P
 autogluon/tabular/models/mitra/_internal/core/get_optimizer.py,sha256=UgGO6lduVZTKZmYAmE207o2Dqs4e3_hyzaoSOQ0iK6A,3412
 autogluon/tabular/models/mitra/_internal/core/get_scheduler.py,sha256=2lzdAxDOYZNq76pmK-FjCOX5MX6cqUSMjqVu8BX9jfY,2238
 autogluon/tabular/models/mitra/_internal/core/prediction_metrics.py,sha256=fai0VnDm0mNjJzx8e1JXdB77PKQsmfbtn8zybD9_qD0,4394
-autogluon/tabular/models/mitra/_internal/core/trainer_finetune.py,sha256=LWw60of990QFYKAmKZJytERjj5_m1sveYyRFqPcb6DE,17527
+autogluon/tabular/models/mitra/_internal/core/trainer_finetune.py,sha256=tI8sN9mv3PtEBdmDxcBgzderZ7YQdtn6MxtOWAc8or8,17908
 autogluon/tabular/models/mitra/_internal/data/__init__.py,sha256=u4ZTvTQNIHqqxilkVqTmYShI2jFMCOyMdv1GRExvtj0,42
 autogluon/tabular/models/mitra/_internal/data/collator.py,sha256=o2F7ODs_eUnV947lCQTx9RugrANidCdiwnZWtdVNJnE,2300
-autogluon/tabular/models/mitra/_internal/data/dataset_finetune.py,sha256=M2QbXjnb5b4CK5qBthWa7bGvsi8Ox8cz_D0u7tBD4Mo,4232
-autogluon/tabular/models/mitra/_internal/data/dataset_split.py,sha256=xpG62WFjg9NTqukKSJx3byq-SFqhxgpIG4jwIl1YuEc,1929
+autogluon/tabular/models/mitra/_internal/data/dataset_finetune.py,sha256=AYxyQ1NJZ3pAp6ny-Y_hqw_4VtyW5X1AABchf7pVsSM,4340
+autogluon/tabular/models/mitra/_internal/data/dataset_split.py,sha256=0uvfyiKrzipde4ZcCDwTE1E3zHelE8xbuNvCeL38J5c,2033
 autogluon/tabular/models/mitra/_internal/data/preprocessor.py,sha256=zx2pWrpDaGSSawPaj7ieRjFOtct_Fyh8LYjo_YtlNG0,13821
 autogluon/tabular/models/mitra/_internal/models/__init__.py,sha256=K0vh5pyrntXp-o7gWNgQ0ZvDbxgeQuRgb6u8ecdjFhA,45
 autogluon/tabular/models/mitra/_internal/models/base.py,sha256=PKpMPT5OT9JFnmYPnhzFUeZPwdNM1e-k97_gW8GZq0Y,468
 autogluon/tabular/models/mitra/_internal/models/embedding.py,sha256=74O6cGWhUyHxg4-wiQwy4sPeDYQze2ekI9H5mLUtSLg,6223
-autogluon/tabular/models/mitra/_internal/models/tab2d.py,sha256=w73QQrXZA7m2fdEPpJDVx-XVZK8xWdc_Q1F38uAZiZA,25690
+autogluon/tabular/models/mitra/_internal/models/tab2d.py,sha256=TorZsQR7LE5QRq2EAq1iT2asLuuAHpgy-PXXrTMxgSs,25743
 autogluon/tabular/models/mitra/_internal/utils/__init__.py,sha256=0mhykAqjMmcEc8Y2od_DMPMk8f66LZHWM7qFdUrPddU,34
 autogluon/tabular/models/mitra/_internal/utils/set_seed.py,sha256=UnXzYfhmfT_tNAofKtLkKpwB9b6HVf9cpI4mKvoBuNM,340
 autogluon/tabular/models/realmlp/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -103,12 +104,12 @@ autogluon/tabular/models/rf/compilers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCe
 autogluon/tabular/models/rf/compilers/native.py,sha256=HhaqQRkVuf9UEEJPsHcdYCmuWBMYtyqRwwB_N2qxG2M,1313
 autogluon/tabular/models/rf/compilers/onnx.py,sha256=pvaZWdl2JJaE2pFU0mFugzhnybePqe0x1-5oLOvogA0,4318
 autogluon/tabular/models/tabicl/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autogluon/tabular/models/tabicl/tabicl_model.py,sha256=bOCOW2E2bcWQRik2gmebKDEzevswQO_3WAF0JVX-Sis,6038
+autogluon/tabular/models/tabicl/tabicl_model.py,sha256=ovytUwFdqpVelEp1cDhHczWQkG72icAnuXTW6GdBi_8,6087
 autogluon/tabular/models/tabm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/tabm/_tabm_internal.py,sha256=fRQ-s5PN94kWqf3LRDen7su_fd-d332YKxdms30FoZM,21066
-autogluon/tabular/models/tabm/rtdl_num_embeddings.py,sha256=omDKJT0MjniUPUnk8tSU-brE8dXIjw27BHFbYc2bswQ,30119
+autogluon/tabular/models/tabm/rtdl_num_embeddings.py,sha256=XssNMaUM0E0G8Grzl_VkVsLt2FcMf3I4cplfvQdVum0,30156
 autogluon/tabular/models/tabm/tabm_model.py,sha256=IQ4RHM1wnf9GHuEa1zDO_yWUPfmh5xUMEVtQ4EFeQRI,10152
-autogluon/tabular/models/tabm/tabm_reference.py,sha256=sZt1LGdifDfJyauVb8wBs9h6lXZJVe0fz0v6oIjXw5A,21908
+autogluon/tabular/models/tabm/tabm_reference.py,sha256=byyP6lcJjA4THbP1VDTgJkj62zyz2S3mEvxWB-kFROw,21944
 autogluon/tabular/models/tabpfnmix/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/tabpfnmix/tabpfnmix_model.py,sha256=7cLjAfstq6Xb-l2DxBdwtSAIanSJN2sMfKPtijDQwXo,16193
 autogluon/tabular/models/tabpfnmix/_internal/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -136,7 +137,7 @@ autogluon/tabular/models/tabpfnmix/_internal/models/foundation/foundation_transf
 autogluon/tabular/models/tabpfnmix/_internal/results/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/tabpfnmix/_internal/results/prediction_metrics.py,sha256=1tRPHyViSSLJ7BkQJi6wai-PwXJ56od86Dy1WWKWZq4,1743
 autogluon/tabular/models/tabpfnv2/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autogluon/tabular/models/tabpfnv2/tabpfnv2_model.py,sha256=dqjDUpIKQ-SIvbeaDVTq1LfmH4iJ1qRVKpb5_ZMM6oE,14296
+autogluon/tabular/models/tabpfnv2/tabpfnv2_model.py,sha256=e3M_JM92ebbOAk4raZSgrWxaVkGlbbMedlHJVmyXfJ4,14570
 autogluon/tabular/models/tabpfnv2/rfpfn/__init__.py,sha256=yE5XAhGxKEFV0JcelZ_JTQZIWGlVEVUQ9a-lxcH_Esc,585
 autogluon/tabular/models/tabpfnv2/rfpfn/configs.py,sha256=lzBY9kKOeBZACVrtRDPHF4ATs9g1rxyNnIs2CMjE20c,1175
 autogluon/tabular/models/tabpfnv2/rfpfn/scoring_utils.py,sha256=uvHsfvnnMdg4tP3_7zAilktkw7nr65LaqfVKXabXAow,6785
@@ -172,7 +173,7 @@ autogluon/tabular/models/xt/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMp
 autogluon/tabular/models/xt/xt_model.py,sha256=qOHJ5h1lHI7uYJfbl0BWm-29R3MNp2WeZB9ptcq5Xis,1003
 autogluon/tabular/predictor/__init__.py,sha256=zCMgjxQlWpDWnr1l1xjBCiK3rWC3N3RoD8UXBnazT74,107
 autogluon/tabular/predictor/interpretable_predictor.py,sha256=5UeKgnMFsfY65tiO3kxfHBPr03lyswLrgdtjPhI0Y7Q,6934
-autogluon/tabular/predictor/predictor.py,sha256=cjszntXs6k5BZMOaLGaMiC1e2sGkCsnXrH9rVI972-0,356548
+autogluon/tabular/predictor/predictor.py,sha256=E4Z_2V0T3tBNuH8dxCm1pDpGkyGDs7rEgDE1SKOy0ow,359075
 autogluon/tabular/registry/__init__.py,sha256=vZpzX4Xve7bfA9crt5LxjgQv9PPfxbi1E1U6Im0Y_xU,93
 autogluon/tabular/registry/_ag_model_registry.py,sha256=Aa-o_KZZiroPBpvZozIBXOlWYvQJN-MVsl_Gl66gkE8,1550
 autogluon/tabular/registry/_model_registry.py,sha256=Rl8Q7BLzaif4hxNxJF20xGE02vrWwh2ZuUaTmA-UJnE,6824
@@ -181,18 +182,18 @@ autogluon/tabular/testing/fit_helper.py,sha256=0eTvPtqM8k8hlOUIHQiwTzik4juTjHQt1
 autogluon/tabular/testing/generate_datasets.py,sha256=nvcAmI-tOh5fwx_ZTx2aRa1n7CsXb96wbR-xqNy1C5w,3884
 autogluon/tabular/testing/model_fit_helper.py,sha256=ZjWpw2nyeFnsrccmkfQtx3qbA8HJx282XX2rwdS-LIs,3808
 autogluon/tabular/trainer/__init__.py,sha256=PW_PGL-tWoQzx3ES2S53bQEZOtsRWTYiM9QdOqsk0dI,38
-autogluon/tabular/trainer/abstract_trainer.py,sha256=8uP3DNJCgmQuDWG0zGJnSNrMCVyMbe8mm11hA8GSP98,232409
+autogluon/tabular/trainer/abstract_trainer.py,sha256=9FiBqOV2h8era6KfydFSqhTlh7RnHkvlvzqsZuij7nE,232527
 autogluon/tabular/trainer/auto_trainer.py,sha256=ZQgQKFT1iHzzun5o5ojdq5pSQmr9ctTkNhe2r9OPOr0,8731
 autogluon/tabular/trainer/model_presets/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/trainer/model_presets/presets.py,sha256=hoWADaOG576Q_XLV1nY_ju1OWi7EJwHay4jjljqt_E0,16546
 autogluon/tabular/trainer/model_presets/presets_distill.py,sha256=MnFC2GJc6RmDBNAGbsO2XMfo3PjR8cUrZoilWW8gTYQ,3295
 autogluon/tabular/tuning/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/tuning/feature_pruner.py,sha256=9iNku8gVbYEkjuKlyITPJDicsNkoraaQOlINQq9iZlQ,6877
-autogluon.tabular-1.3.2b20250722.dist-info/LICENSE,sha256=CeipvOyAZxBGUsFoaFqwkx54aPnIKEtm9a5u2uXxEws,10142
-autogluon.tabular-1.3.2b20250722.dist-info/METADATA,sha256=wtSNYXyA4ratSdv6547JHxGPrJOyU8Xq-8j8hFhZcho,14875
-autogluon.tabular-1.3.2b20250722.dist-info/NOTICE,sha256=7nPQuj8Kp-uXsU0S5so3-2dNU5EctS5hDXvvzzehd7E,114
-autogluon.tabular-1.3.2b20250722.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-autogluon.tabular-1.3.2b20250722.dist-info/namespace_packages.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
-autogluon.tabular-1.3.2b20250722.dist-info/top_level.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
-autogluon.tabular-1.3.2b20250722.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
-autogluon.tabular-1.3.2b20250722.dist-info/RECORD,,
+autogluon.tabular-1.4.0b20250724.dist-info/LICENSE,sha256=CeipvOyAZxBGUsFoaFqwkx54aPnIKEtm9a5u2uXxEws,10142
+autogluon.tabular-1.4.0b20250724.dist-info/METADATA,sha256=N8GqPqMCIoGwOwmffPt1scTY3FPCZdP3f3uds0Nm8PI,16071
+autogluon.tabular-1.4.0b20250724.dist-info/NOTICE,sha256=7nPQuj8Kp-uXsU0S5so3-2dNU5EctS5hDXvvzzehd7E,114
+autogluon.tabular-1.4.0b20250724.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+autogluon.tabular-1.4.0b20250724.dist-info/namespace_packages.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
+autogluon.tabular-1.4.0b20250724.dist-info/top_level.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
+autogluon.tabular-1.4.0b20250724.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
+autogluon.tabular-1.4.0b20250724.dist-info/RECORD,,