autogluon.tabular 1.2.1b20250116__py3-none-any.whl → 1.2.1b20250130__py3-none-any.whl

autogluon/tabular/experimental/_tabular_classifier.py

@@ -1,7 +1,5 @@
 from __future__ import annotations
 
-from typing import List
-
 import pandas as pd
 from sklearn.base import BaseEstimator, ClassifierMixin
 from sklearn.utils.multiclass import unique_labels
@@ -18,7 +16,7 @@ class TabularClassifier(BaseEstimator, ClassifierMixin, ScikitMixin):
         self,
         eval_metric: str | Scorer = None,
         time_limit: float = None,
-        presets: List[str] | str = None,
+        presets: list[str] | str = None,
         hyperparameters: dict | str = None,
         path: str = None,
         verbosity: int = 2,

autogluon/tabular/experimental/_tabular_regressor.py

@@ -1,7 +1,5 @@
 from __future__ import annotations
 
-from typing import List
-
 import pandas as pd
 from sklearn.base import BaseEstimator, RegressorMixin
 from sklearn.utils.validation import check_array, check_is_fitted, check_X_y
@@ -17,7 +15,7 @@ class TabularRegressor(BaseEstimator, RegressorMixin, ScikitMixin):
         self,
         eval_metric: str | Scorer = None,
         time_limit: float = None,
-        presets: List[str] | str = None,
+        presets: list[str] | str = None,
         hyperparameters: dict | str = None,
         path: str = None,
         verbosity: int = 2,

autogluon/tabular/models/_utils/rapids_utils.py

@@ -4,7 +4,7 @@ from autogluon.common.utils.resource_utils import ResourceManager
 
 
 class RapidsModelMixin:
-    """Mixin class for methods re-used across RAPIDS models"""
+    """Mixin class for methods reused across RAPIDS models"""
 
     # FIXME: Efficient OOF doesn't work in RAPIDS
     @classmethod

autogluon/tabular/models/catboost/callbacks.py

@@ -39,37 +39,65 @@ class MemoryCheckCallback:
                 return False
         return True
 
-    def memory_check(self, iter) -> bool:
-        """Checks if memory usage is unsafe. If so, then returns True to signal the model to stop training early."""
+    def memory_check(self, iter: int) -> bool:
+        """
+        Checks if memory usage is unsafe. If so, signals the model to stop training early.
+
+        Parameters
+        ----------
+        iter: int
+            The current training iteration.
+
+        Returns
+        -------
+        bool: True if training should stop due to memory constraints, False otherwise.
+        """
         available_bytes = ResourceManager.get_available_virtual_mem()
         cur_rss = ResourceManager.get_memory_rss()
 
+        # Update initial memory usage if current usage is lower
         if cur_rss < self.init_mem_rss:
             self.init_mem_rss = cur_rss
-        estimated_model_size_mb = (cur_rss - self.init_mem_rss) >> 20
-        available_mb = available_bytes >> 20
-        model_size_memory_ratio = estimated_model_size_mb / available_mb
 
+        # Convert memory values to MB
+        estimated_model_size_mb = (cur_rss - self.init_mem_rss) / (1024 ** 2)
+        available_mb = available_bytes / (1024 ** 2)
+
+        model_size_memory_ratio = estimated_model_size_mb / available_mb
         early_stop = False
+
         if model_size_memory_ratio > 1.0:
-            logger.warning(f"Warning: Large model size may cause OOM error if training continues")
+            logger.warning(
+                f"Iteration {iter}: Model size exceeds available memory. "
+                f"Available memory: {available_mb:.2f} MB, "
+                f"Estimated model size: {estimated_model_size_mb:.2f} MB."
+            )
             early_stop = True
 
-        if available_mb < 512:  # Less than 500 MB
-            logger.warning(f"Warning: Low available memory may cause OOM error if training continues")
+        if available_mb < 512:  # Less than 512 MB
+            logger.warning(
+                f"Iteration {iter}: Low available memory (<512 MB). "
+                f"Available memory: {available_mb:.2f} MB, "
+                f"Estimated model size: {estimated_model_size_mb:.2f} MB."
+            )
             early_stop = True
 
         if early_stop:
             logger.warning(
-                "Warning: Early stopped model prior to optimal result to avoid OOM error. " "Please increase available memory to avoid subpar model quality."
+                "Early stopping model prior to optimal result to avoid OOM error. "
+                "Please increase available memory to avoid subpar model quality."
             )
-            logger.warning(f"Available Memory: {available_mb} MB, Estimated Model size: {estimated_model_size_mb} MB")
             return True
-        elif self.verbose or (model_size_memory_ratio > 0.25):
-            logging.debug(f"Available Memory: {available_mb} MB, Estimated Model size: {estimated_model_size_mb} MB")
+        elif self.verbose or model_size_memory_ratio > 0.25:
+            logger.debug(
+                f"Iteration {iter}: "
+                f"Available memory: {available_mb:.2f} MB, "
+                f"Estimated model size: {estimated_model_size_mb:.2f} MB."
+            )
 
+        # Adjust memory check frequency based on model size
         if model_size_memory_ratio > 0.5:
-            self._cur_period = 1  # Increase rate of memory check if model gets large enough to cause OOM potentially
+            self._cur_period = 1  # Increase frequency of memory checks
         elif iter > self.period:
             self._cur_period = self.period
 

autogluon/tabular/models/fastainn/tabular_nn_fastai.py

@@ -8,7 +8,7 @@ import warnings
 from builtins import classmethod
 from functools import partial
 from pathlib import Path
-from typing import Dict, Union
+from typing import Union
 
 import numpy as np
 import pandas as pd
@@ -628,7 +628,7 @@ class NNFastAiTabularModel(AbstractModel):
         """Choose which backend(Ray or Custom) to use for hpo"""
         return RAY_BACKEND
 
-    def _get_maximum_resources(self) -> Dict[str, Union[int, float]]:
+    def _get_maximum_resources(self) -> dict[str, Union[int, float]]:
         # fastai model trains slower when utilizing virtual cores and this issue scale up when the number of cpu cores increases
         return {"num_cpus": ResourceManager.get_cpu_count_psutil(logical=False)}
 

autogluon/tabular/models/lgb/lgb_utils.py

@@ -1,7 +1,7 @@
 import copy
 import os
 import time
-from typing import List, Optional
+from typing import Optional
 
 import numpy as np
 from pandas import DataFrame, Series
@@ -137,7 +137,7 @@ def train_lgb_model(early_stopping_callback_kwargs=None, **train_params):
 class QuantileBooster:
     """Wrapper that trains a separate LGBM Booster for each quantile level."""
 
-    def __init__(self, quantile_levels: List[float], early_stopping_callback_kwargs: Optional[dict] = None):
+    def __init__(self, quantile_levels: list[float], early_stopping_callback_kwargs: Optional[dict] = None):
         if quantile_levels is None:
             raise AssertionError
         if not all(0 < q < 1 for q in quantile_levels):

autogluon/tabular/models/tab_transformer/modified_transformer.py

@@ -6,7 +6,7 @@ The modification allows the option of fixing the attention map
 """
 
 import math
-from typing import Optional, Tuple
+from typing import Optional
 
 import torch
 import torch.nn.functional as F
@@ -54,7 +54,7 @@ def multi_head_attention_forward(
     static_k=None,  # type: Optional[Tensor]
     static_v=None,  # type: Optional[Tensor]
 ):
-    # type: (...) -> Tuple[Tensor, Optional[Tensor]]
+    # type: (...) -> tuple[Tensor, Optional[Tensor]]
     """
     Args:
         query, key, value: map a query and a set of key-value pairs to an output.
@@ -288,7 +288,7 @@ class MultiheadAttention(Module):
         super().__setstate__(state)
 
     def forward(self, query, key, value, key_padding_mask=None, need_weights=True, attn_mask=None):
-        # type: (Tensor, Tensor, Tensor, Optional[Tensor], bool, Optional[Tensor]) -> Tuple[Tensor, Optional[Tensor]]
+        # type: (Tensor, Tensor, Tensor, Optional[Tensor], bool, Optional[Tensor]) -> tuple[Tensor, Optional[Tensor]]
         """
         Args:
             query, key, value: map a query and a set of key-value pairs to an output.

autogluon/tabular/models/tab_transformer/pretexts.py

@@ -111,8 +111,8 @@ class BERTPretext(nn.Module):
             assert torch.all(cat_feats[extra_replace] == orig_cat_feats[extra_replace]).item() is True
             extra_plus1 = cat_feats[extra_replace] + 1
             extra_minus1 = cat_feats[extra_replace] - 1
-            extra_zero_padd_idx = extra_minus1 == 0
-            extra_minus1[extra_zero_padd_idx] = extra_plus1[extra_zero_padd_idx]
+            extra_zero_pad_idx = extra_minus1 == 0
+            extra_minus1[extra_zero_pad_idx] = extra_plus1[extra_zero_pad_idx]
 
             cat_feats[extra_replace] = extra_minus1
             assert torch.all(~(cat_feats[extra_replace] == orig_cat_feats[extra_replace])).item() is True
@@ -135,8 +135,8 @@ class BERTPretext(nn.Module):
             assert torch.all(cat_feats[extra_replace] == orig_cat_feats[extra_replace]).item() is True
             extra_plus1 = cat_feats[extra_replace] + 1
             extra_minus1 = cat_feats[extra_replace] - 1
-            extra_zero_padd_idx = extra_minus1 == 0
-            extra_minus1[extra_zero_padd_idx] = extra_plus1[extra_zero_padd_idx]
+            extra_zero_pad_idx = extra_minus1 == 0
+            extra_minus1[extra_zero_pad_idx] = extra_plus1[extra_zero_pad_idx]
 
             cat_feats[extra_replace] = extra_minus1
             assert torch.all(~(cat_feats[extra_replace] == orig_cat_feats[extra_replace])).item() is True

autogluon/tabular/models/tab_transformer/tab_transformer_encoder.py

@@ -7,7 +7,7 @@ import re
 from collections import Counter
 from datetime import date, datetime
 from functools import partial
-from typing import Iterable, List, Union
+from typing import Iterable, Union
 
 import numpy as np
 import pandas as pd
@@ -514,7 +514,7 @@ class EmbeddingInitializer(nn.Module):
         :param drop_whole_embeddings:
             If True, dropout pretends the embedding was a missing value. If false, dropout sets embed features to 0
         :param one_hot:
-            If True, one-hot encode variables whose cardinality is < max_emb_dim. Also, set reqiures_grad = False
+            If True, one-hot encode variables whose cardinality is < max_emb_dim. Also, set requires_grad = False
         :param out_dim:
             If None, return the embedding straight from self.embed.  If another dimension, put the embedding through a
             Linear layer to make it size (batch x out_dim).
@@ -624,7 +624,7 @@ def add_datepart(df: DataFrame, field_name: str, prefix: str = None, drop: bool
     return df
 
 
-def cyclic_dt_feat_names(time: bool = True, add_linear: bool = False) -> List[str]:
+def cyclic_dt_feat_names(time: bool = True, add_linear: bool = False) -> list[str]:
     "Return feature names of date/time cycles as produced by `cyclic_dt_features`."
     fs = ["cos", "sin"]
     attr = [f"{r}_{f}" for r in "weekday day_month month_year day_year".split() for f in fs]
@@ -635,7 +635,7 @@ def cyclic_dt_feat_names(time: bool = True, add_linear: bool = False) -> List[st
     return attr
 
 
-def cyclic_dt_features(d: Union[date, datetime], time: bool = True, add_linear: bool = False) -> List[float]:
+def cyclic_dt_features(d: Union[date, datetime], time: bool = True, add_linear: bool = False) -> list[float]:
     "Calculate the cos and sin of date/time cycles."
     tt, fs = d.timetuple(), [np.cos, np.sin]
     day_year, days_month = tt.tm_yday, calendar.monthrange(d.year, d.month)[1]

autogluon/tabular/models/tabular_nn/torch/tabular_torch_dataset.py

@@ -136,7 +136,7 @@ class TabularTorchDataset(torch.utils.data.IterableDataset):
         load the dataset.
 
         Returns a tuple containing (vector_features, embed_features, label).
-        The length of the tuple depends on `has_vector_features` and `has_embed_feautures` attribute.
+        The length of the tuple depends on `has_vector_features` and `has_embed_features` attribute.
         """
         idxarray = np.arange(self.num_examples)
         if self.shuffle:
@@ -154,7 +154,7 @@ class TabularTorchDataset(torch.utils.data.IterableDataset):
                 idx = idxarray[idx]
 
             # Generate a tuple that contains (vector_features, embed_features, label).
-            # The length of the tuple depends on `has_vector_features`, `has_embed_feautures`, and
+            # The length of the tuple depends on `has_vector_features`, `has_embed_features`, and
             # whether the label has been provided.
             output_list = []
             if self.has_vector_features:

autogluon/tabular/predictor/_deprecated_methods.py

@@ -1,7 +1,5 @@
 from __future__ import annotations
 
-from typing import Dict, List
-
 import numpy as np
 import pandas as pd
 
@@ -12,17 +10,17 @@ class TabularPredictorDeprecatedMixin:
     """Contains deprecated methods from TabularPredictor that shouldn't show up in API documentation."""
 
     @Deprecated(min_version_to_warn="0.8.3", min_version_to_error="1.2", version_to_remove="1.2", new="persist")
-    def persist_models(self, *args, **kwargs) -> List[str]:
+    def persist_models(self, *args, **kwargs) -> list[str]:
         """Deprecated method. Use `persist` instead."""
         return self.persist(*args, **kwargs)
 
     @Deprecated(min_version_to_warn="0.8.3", min_version_to_error="1.2", version_to_remove="1.2", new="unpersist")
-    def unpersist_models(self, *args, **kwargs) -> List[str]:
+    def unpersist_models(self, *args, **kwargs) -> list[str]:
         """Deprecated method. Use `unpersist` instead."""
         return self.unpersist(*args, **kwargs)
 
     @Deprecated(min_version_to_warn="0.8.3", min_version_to_error="1.2", version_to_remove="1.2", new="model_names")
-    def get_model_names(self, *args, **kwargs) -> List[str]:
+    def get_model_names(self, *args, **kwargs) -> list[str]:
         """Deprecated method. Use `model_names` instead."""
         return self.model_names(*args, **kwargs)
 
@@ -37,7 +35,7 @@ class TabularPredictorDeprecatedMixin:
         return self.predict_from_proba(*args, **kwargs)
 
     @Deprecated(min_version_to_warn="0.8.3", min_version_to_error="1.2", version_to_remove="1.2", new="model_refit_map")
-    def get_model_full_dict(self, *args, **kwargs) -> Dict[str, str]:
+    def get_model_full_dict(self, *args, **kwargs) -> dict[str, str]:
         """Deprecated method. Use `model_refit_map` instead."""
         return self.model_refit_map(*args, **kwargs)
 
@@ -62,6 +60,6 @@ class TabularPredictorDeprecatedMixin:
         return self.disk_usage()
 
     @Deprecated(min_version_to_warn="0.8.3", min_version_to_error="1.2", version_to_remove="1.2", new="model_names(persisted=True)")
-    def get_model_names_persisted(self) -> List[str]:
+    def get_model_names_persisted(self) -> list[str]:
         """Deprecated method. Use `model_names(persisted=True)` instead."""
         return self.model_names(persisted=True)

autogluon/tabular/predictor/predictor.py

@@ -9,7 +9,7 @@ import pprint
 import shutil
 import time
 import warnings
-from typing import overload, Any, Dict, List, Literal, Optional, Tuple, Union
+from typing import overload, Any, Literal, Optional, Union
 
 import networkx as nx
 import numpy as np
@@ -224,7 +224,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         )
         self._learner_type = type(self._learner)
         self._trainer: AbstractTrainer = None
-        self._sub_fits: List[str] = []
+        self._sub_fits: list[str] = []
         self._stacked_overfitting_occurred: bool | None = None
         self._fit_strategy = None
 
@@ -271,7 +271,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         return self._learner.label_cleaner.inv_map
 
     @property
-    def quantile_levels(self) -> List[float]:
+    def quantile_levels(self) -> list[float]:
         return self._learner.quantile_levels
 
     @property
@@ -280,7 +280,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         return self._learner.eval_metric
 
     @property
-    def original_features(self) -> List[str]:
+    def original_features(self) -> list[str]:
         """Original features user passed in to fit before processing"""
         self._assert_is_fit()
         return self._learner.original_features
@@ -400,7 +400,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         train_data: pd.DataFrame | str,
         tuning_data: pd.DataFrame | str = None,
         time_limit: float = None,
-        presets: List[str] | str = None,
+        presets: list[str] | str = None,
         hyperparameters: dict | str = None,
         feature_metadata="infer",
         infer_limit: float = None,
@@ -414,7 +414,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         num_gpus: int | str = "auto",
         fit_strategy: Literal["sequential", "parallel"] = "sequential",
         memory_limit: float | str = "auto",
-        callbacks: List[AbstractCallback] = None,
+        callbacks: list[AbstractCallback] = None,
         **kwargs,
     ) -> "TabularPredictor":
         """
@@ -531,7 +531,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             Advanced functionality: Bring your own model / Custom model support
                 AutoGluon fully supports custom models. For a detailed tutorial on creating and using custom models with AutoGluon, refer to https://auto.gluon.ai/stable/tutorials/tabular/advanced/tabular-custom-model.html
             Advanced functionality: Custom stack levels
-                By default, AutoGluon re-uses the same models and model hyperparameters at each level during stack ensembling.
+                By default, AutoGluon reuses the same models and model hyperparameters at each level during stack ensembling.
                 To customize this behaviour, create a hyperparameters dictionary separately for each stack level, and then add them as values to a new dictionary, with keys equal to the stack level.
                     Example: `hyperparameters = {1: {'RF': rf_params1}, 2: {'CAT': [cat_params1, cat_params2], 'NN_TORCH': {}}}`
                     This will result in a stack ensemble that has one custom random forest in level 1 followed by two CatBoost models with custom hyperparameters and a default neural network in level 2, for a total of 4 models.
@@ -737,7 +737,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             setting the memory limit (and any other resources) on systems with shared resources that are controlled by the operating system (e.g., SLURM and
             cgroups). Otherwise, AutoGluon might wrongly assume more resources are available for fitting a model than the operating system allows,
             which can result in model training failing or being very inefficient.
-        callbacks : List[AbstractCallback], default = None
+        callbacks : list[AbstractCallback], default = None
             :::{warning}
             Callbacks are an experimental feature and may change in future releases without warning.
             Callback support is preliminary and targeted towards developers.
@@ -750,7 +750,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             If None, no callback objects will be used.
 
             [Note] Callback objects can be mutated in-place by the fit call if they are stateful.
-            Ensure that you avoid re-using a mutated callback object between multiple fit calls.
+            Ensure that you avoid reusing a mutated callback object between multiple fit calls.
 
             [Note] Callback objects are deleted from trainer at the end of the fit call. They will not impact operations such as `refit_full` or `fit_extra`.
         **kwargs :
@@ -1779,9 +1779,9 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
     # TODO: Consider adding infer_limit to fit_extra
     def fit_extra(
         self,
-        hyperparameters: str | Dict[str, Any],
+        hyperparameters: str | dict[str, Any],
         time_limit: float = None,
-        base_model_names: List[str] = None,
+        base_model_names: list[str] = None,
         fit_weighted_ensemble: bool = True,
         fit_full_last_level_weighted_ensemble: bool = True,
         full_weighted_ensemble_additionally: bool = False,
@@ -1804,7 +1804,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             dictionary is relative, not absolute.
         time_limit : int, default = None
             Refer to argument documentation in :meth:`TabularPredictor.fit`.
-        base_model_names : List[str], default = None
+        base_model_names : list[str], default = None
             The names of the models to use as base models for this fit call.
             Base models will provide their out-of-fold predictions as additional features to the models in `hyperparameters`.
             If specified, all models trained will be stack ensembles.
@@ -2294,7 +2294,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         # TODO: Consider making calculating this information easier, such as keeping track of meta-info from the latest/original fit call.
         #  Currently we use `stack_name == core` to figure out the number of stack levels, but this is somewhat brittle.
         if "num_stack_levels" not in fit_extra_kwargs and not was_fit:
-            models_core: List[str] = [m for m, stack_name in self._trainer.get_models_attribute_dict(attribute="stack_name").items() if stack_name == "core"]
+            models_core: list[str] = [m for m, stack_name in self._trainer.get_models_attribute_dict(attribute="stack_name").items() if stack_name == "core"]
             num_stack_levels = max(self._trainer.get_models_attribute_dict(attribute="level", models=models_core).values()) - 1
             fit_extra_kwargs["num_stack_levels"] = num_stack_levels
         if is_labeled:
@@ -2785,7 +2785,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             display=display,
         )
 
-    def learning_curves(self) -> Tuple[dict, dict]:
+    def learning_curves(self) -> tuple[dict, dict]:
         """
         Retrieves learning curves generated during predictor.fit().
         Will not work if the learning_curves flag was not set during training.
@@ -2892,7 +2892,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
     def predict_proba_multi(
         self,
         data: pd.DataFrame = None,
-        models: List[str] = None,
+        models: list[str] = None,
         as_pandas: bool = True,
         as_multiclass: bool = True,
         transform_features: bool = True,
@@ -2924,7 +2924,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
                 for m in models:
                     predict_proba_dict[m] = predictor.predict_proba_oof(model=m)
                 ```
-        models : List[str], default = None
+        models : list[str], default = None
             The list of models to get predictions for.
             If None, all models that can infer are used.
         as_pandas : bool, default = True
@@ -2972,7 +2972,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
     def predict_multi(
         self,
         data: pd.DataFrame = None,
-        models: List[str] = None,
+        models: list[str] = None,
         as_pandas: Literal[True] = True,
         transform_features: bool = True,
         inverse_transform: bool = True,
@@ -2983,7 +2983,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
     def predict_multi(
         self,
         data: pd.DataFrame = None,
-        models: List[str] = None,
+        models: list[str] = None,
         *,
         as_pandas: Literal[False],
         transform_features: bool = True,
@@ -2994,7 +2994,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
     def predict_multi(
         self,
         data: pd.DataFrame = None,
-        models: List[str] = None,
+        models: list[str] = None,
         as_pandas: bool = True,
         transform_features: bool = True,
         inverse_transform: bool = True,
@@ -3027,7 +3027,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
                 for m in models:
                     predict_dict[m] = predictor.predict_oof(model=m)
                 ```
-        models : List[str], default = None
+        models : list[str], default = None
             The list of models to get predictions for.
             If None, all models that can infer are used.
         as_pandas : bool, default = True
@@ -3246,7 +3246,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             Valid models are listed in this `predictor` by calling `predictor.model_names()`.
             Specifying a `refit_full` model will cause an exception if `data=None`.
             `base_models=None` is a requirement when specifying `model`.
-        base_models: List[str], default = None
+        base_models: list[str], default = None
             List of model names to use as base_models for a hypothetical stacker model when generating input features.
             If `None`, then only return generically preprocessed features prior to any model fitting.
             Valid models are listed in this `predictor` by calling `predictor.model_names()`.
@@ -3516,7 +3516,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
                 raise ValueError(f'Unknown compiler_configs preset: "{compiler_configs}"')
         self._trainer.compile(model_names=models, with_ancestors=with_ancestors, compiler_configs=compiler_configs)
 
-    def persist(self, models="best", with_ancestors=True, max_memory=0.4) -> List[str]:
+    def persist(self, models="best", with_ancestors=True, max_memory=0.4) -> list[str]:
         """
         Persist models in memory for reduced inference latency. This is particularly important if the models are being used for online-inference where low latency is critical.
         If models are not persisted in memory, they are loaded from disk every time they are asked to make predictions.
@@ -3552,7 +3552,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
                     raise ValueError(f"Invalid models specified. The following models do not exist:\n\t{invalid_models}\nValid models:\n\t{valid_models}")
             raise e
 
-    def unpersist(self, models="all") -> List[str]:
+    def unpersist(self, models="all") -> list[str]:
         """
         Unpersist models in memory for reduced memory usage.
         If models are not persisted in memory, they are loaded from disk every time they are asked to make predictions.
@@ -3576,14 +3576,14 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
     #  refit_full doesn't account for user-specified resources at fit time, nor does it allow them to specify for refit.
     def refit_full(
         self,
-        model: str | List[str] = "all",
+        model: str | list[str] = "all",
         set_best_to_refit_full: bool = True,
         train_data_extra: pd.DataFrame = None,
         num_cpus: int | str = "auto",
         num_gpus: int | str = "auto",
         fit_strategy: Literal["auto", "sequential", "parallel"] = "auto",
         **kwargs,
-    ) -> Dict[str, str]:
+    ) -> dict[str, str]:
         """
         Retrain model on all of the data (training + validation).
         For bagged models:
@@ -3607,7 +3607,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
 
         Parameters
         ----------
-        model : str | List[str], default = 'all'
+        model : str | list[str], default = 'all'
             Model name of model(s) to refit.
                 If 'all' then all models are refitted.
                 If 'best' then the model with the highest validation score is refit.
@@ -3759,7 +3759,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         if save_trainer:
             self._trainer.save()
 
-    def model_refit_map(self, inverse=False) -> Dict[str, str]:
+    def model_refit_map(self, inverse=False) -> dict[str, str]:
         """
         Returns a dictionary of original model name -> refit full model name.
         Empty unless `refit_full=True` was set during fit or `predictor.refit_full()` was called.
@@ -3915,7 +3915,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         data: pd.DataFrame | str | None = None,
         metric: str | Scorer | None = None,
         model: str = "best",
-        decision_thresholds: int | List[float] = 25,
+        decision_thresholds: int | list[float] = 25,
         secondary_decision_thresholds: int | None = 19,
         subsample_size: int | None = 1000000,
         verbose: bool = True,
@@ -3942,7 +3942,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         model : str, default = 'best'
             The model to use prediction probabilities of when calibrating the threshold.
             If 'best', will use `predictor.model_best`.
-        decision_thresholds : int | List[float], default = 25
+        decision_thresholds : int | list[float], default = 25
             The number of decision thresholds on either side of `0.5` to search.
             The default of 25 will result in 51 searched thresholds: [0.00, 0.02, 0.04, ..., 0.48, 0.50, 0.52, ..., 0.96, 0.98, 1.00]
             Alternatively, a list of decision thresholds can be passed and only the thresholds in the list will be searched.
@@ -4338,9 +4338,9 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         stack_name: str = None,
         level: int = None,
         can_infer: bool = None,
-        models: List[str] = None,
+        models: list[str] = None,
         persisted: bool = None,
-    ) -> List[str]:
+    ) -> list[str]:
         """
         Returns the list of model names trained in this `predictor` object.
 
@@ -4352,7 +4352,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             If specified, returns only models at the given stack level.
         can_infer: bool, default = None
             If specified, returns only models that can/cannot infer on new data.
-        models: List[str], default = None
+        models: list[str], default = None
             The list of model names to consider.
             If None, considers all models.
         persisted: bool, default = None
@@ -4874,7 +4874,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         return predictor
 
     @classmethod
-    def load_log(cls, predictor_path: str = None, log_file_path: Optional[str] = None) -> List[str]:
+    def load_log(cls, predictor_path: str = None, log_file_path: Optional[str] = None) -> list[str]:
         """
         Load log files of a predictor
 
@@ -4890,7 +4890,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
 
         Returns
         -------
-        List[str]
+        list[str]
             A list containing lines of the log file
         """
         file_path = log_file_path
@@ -5049,7 +5049,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         )
 
     @staticmethod
-    def _sanitize_dynamic_stacking_kwargs(kwargs: dict) -> Tuple[dict, List[str]]:
+    def _sanitize_dynamic_stacking_kwargs(kwargs: dict) -> tuple[dict, list[str]]:
         ds_kwargs_key = "ds_args"
         ds_args = dict(
             validation_procedure="holdout",
@@ -5069,9 +5069,9 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             kwargs[ds_kwargs_key] = copy.deepcopy(kwargs[ds_kwargs_key])
             ds_args.update(kwargs[ds_kwargs_key])
 
-        key_missmatch = set(ds_args.keys()) - allowed_kes
-        if key_missmatch:
-            raise ValueError(f"Got invalid keys for `ds_args`. Allowed: {allowed_kes}. Got: {key_missmatch}")
+        key_mismatch = set(ds_args.keys()) - allowed_kes
+        if key_mismatch:
+            raise ValueError(f"Got invalid keys for `ds_args`. Allowed: {allowed_kes}. Got: {key_mismatch}")
         if ("validation_procedure" in ds_args) and (
             (not isinstance(ds_args["validation_procedure"], str)) or (ds_args["validation_procedure"] not in ["holdout", "cv"])
         ):
@@ -5096,7 +5096,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         kwargs[ds_kwargs_key] = ds_args
         return kwargs, [ds_kwargs_key]
 
-    def _validate_fit_extra_kwargs(self, kwargs: dict, extra_valid_keys: List[str] | None = None):
+    def _validate_fit_extra_kwargs(self, kwargs: dict, extra_valid_keys: list[str] | None = None):
         fit_extra_kwargs_default = self._fit_extra_kwargs_dict()
 
         allowed_kwarg_names = list(fit_extra_kwargs_default.keys())
@@ -5131,7 +5131,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
 
         return kwargs_sanitized
 
-    def _prune_data_features(self, train_features: list, other_features: list, is_labeled: bool) -> Tuple[list, list]:
+    def _prune_data_features(self, train_features: list, other_features: list, is_labeled: bool) -> tuple[list, list]:
         """
         Removes certain columns from the provided datasets that do not contain predictive features.
 
@@ -5161,7 +5161,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         tuning_data: str | pd.DataFrame | None = None,
         test_data: str | pd.DataFrame | None = None,
         unlabeled_data: str | pd.DataFrame | None = None,
-    ) -> Tuple[pd.DataFrame, pd.DataFrame | None, pd.DataFrame | None]:
+    ) -> tuple[pd.DataFrame, pd.DataFrame | None, pd.DataFrame | None]:
         if isinstance(train_data, str):
             train_data = TabularDataset(train_data)
         if tuning_data is not None and isinstance(tuning_data, str):
@@ -5305,7 +5305,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
             )
 
     @staticmethod
-    def _validate_infer_limit(infer_limit: float, infer_limit_batch_size: int) -> Tuple[float, int]:
+    def _validate_infer_limit(infer_limit: float, infer_limit_batch_size: int) -> tuple[float, int]:
         if infer_limit_batch_size is not None:
             if not isinstance(infer_limit_batch_size, int):
                 raise ValueError(f"infer_limit_batch_size must be type int, but was instead type {type(infer_limit_batch_size)}")
@@ -5629,7 +5629,7 @@ class TabularPredictor(TabularPredictorDeprecatedMixin):
         else:
             _validate_hyperparameters_util(params=hyperparameters)
 
-    def _sanitize_pseudo_data(self, pseudo_data: pd.DataFrame, name="pseudo_data") -> Tuple[pd.DataFrame, pd.Series, pd.Series]:
+    def _sanitize_pseudo_data(self, pseudo_data: pd.DataFrame, name="pseudo_data") -> tuple[pd.DataFrame, pd.Series, pd.Series]:
         assert isinstance(pseudo_data, pd.DataFrame)
         if self.label not in pseudo_data.columns:
             raise ValueError(f"'{name}' does not contain the labeled column.")

autogluon/tabular/trainer/auto_trainer.py

@@ -1,9 +1,8 @@
 import logging
-from typing import Dict, List
 
 from autogluon.core.models import AbstractModel
-from autogluon.core.trainer.abstract_trainer import AbstractTrainer
 from autogluon.core.utils import generate_train_test_split
+from autogluon.core.trainer.abstract_trainer import AbstractTabularTrainer
 
 from ..models.lgb.lgb_model import LGBModel
 from .model_presets.presets import MODEL_TYPES, get_preset_models
@@ -13,7 +12,7 @@ logger = logging.getLogger(__name__)
 
 
 # This Trainer handles model training details
-class AutoTrainer(AbstractTrainer):
+class AutoTrainer(AbstractTabularTrainer):
     def construct_model_templates(self, hyperparameters, **kwargs):
         path = kwargs.pop("path", self.path)
         problem_type = kwargs.pop("problem_type", self.problem_type)
@@ -58,7 +57,7 @@ class AutoTrainer(AbstractTrainer):
         infer_limit_batch_size=None,
         use_bag_holdout=False,
         groups=None,
-        callbacks: List[callable] = None,
+        callbacks: list[callable] = None,
         **kwargs,
     ):
         for key in kwargs:
@@ -173,7 +172,7 @@ class AutoTrainer(AbstractTrainer):
     def _get_default_proxy_model_class(self):
         return LGBModel
 
-    def compile(self, model_names="all", with_ancestors=False, compiler_configs: dict = None) -> List[str]:
+    def compile(self, model_names="all", with_ancestors=False, compiler_configs: dict = None) -> list[str]:
         """Ensures that compiler_configs maps to the correct models if the user specified the same keys as in hyperparameters such as RT, XT, etc."""
         if compiler_configs is not None:
             model_types_map = self._get_model_types_map()
@@ -186,5 +185,5 @@ class AutoTrainer(AbstractTrainer):
             compiler_configs = compiler_configs_new
         return super().compile(model_names=model_names, with_ancestors=with_ancestors, compiler_configs=compiler_configs)
 
-    def _get_model_types_map(self) -> Dict[str, AbstractModel]:
+    def _get_model_types_map(self) -> dict[str, AbstractModel]:
         return MODEL_TYPES

autogluon/tabular/version.py

@@ -1,4 +1,4 @@
 """This is the autogluon version file."""
 
-__version__ = "1.2.1b20250116"
+__version__ = "1.2.1b20250130"
 __lite__ = False

{autogluon.tabular-1.2.1b20250116.dist-info → autogluon.tabular-1.2.1b20250130.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: autogluon.tabular
-Version: 1.2.1b20250116
+Version: 1.2.1b20250130
 Summary: Fast and Accurate ML in 3 Lines of Code
 Home-page: https://github.com/autogluon/autogluon
 Author: AutoGluon Community
@@ -34,24 +34,26 @@ Classifier: Topic :: Scientific/Engineering :: Information Analysis
 Classifier: Topic :: Scientific/Engineering :: Image Recognition
 Requires-Python: >=3.9, <3.13
 Description-Content-Type: text/markdown
+License-File: ../LICENSE
+License-File: ../NOTICE
 Requires-Dist: numpy<2.1.4,>=1.25.0
 Requires-Dist: scipy<1.16,>=1.5.4
 Requires-Dist: pandas<2.3.0,>=2.0.0
 Requires-Dist: scikit-learn<1.5.3,>=1.4.0
 Requires-Dist: networkx<4,>=3.0
-Requires-Dist: autogluon.core==1.2.1b20250116
-Requires-Dist: autogluon.features==1.2.1b20250116
+Requires-Dist: autogluon.core==1.2.1b20250130
+Requires-Dist: autogluon.features==1.2.1b20250130
 Provides-Extra: all
+Requires-Dist: autogluon.core[all]==1.2.1b20250130; extra == "all"
+Requires-Dist: spacy<3.8; extra == "all"
+Requires-Dist: lightgbm<4.6,>=4.0; extra == "all"
+Requires-Dist: fastai<2.8,>=2.3.1; extra == "all"
 Requires-Dist: numpy<2.0.0,>=1.25; extra == "all"
 Requires-Dist: xgboost<2.2,>=1.6; extra == "all"
+Requires-Dist: torch<2.6,>=2.2; extra == "all"
 Requires-Dist: einops<0.9,>=0.7; extra == "all"
-Requires-Dist: autogluon.core[all]==1.2.1b20250116; extra == "all"
 Requires-Dist: catboost<1.3,>=1.2; extra == "all"
-Requires-Dist: torch<2.6,>=2.2; extra == "all"
-Requires-Dist: spacy<3.8; extra == "all"
-Requires-Dist: fastai<2.8,>=2.3.1; extra == "all"
 Requires-Dist: huggingface-hub[torch]; extra == "all"
-Requires-Dist: lightgbm<4.6,>=4.0; extra == "all"
 Provides-Extra: catboost
 Requires-Dist: numpy<2.0.0,>=1.25; extra == "catboost"
 Requires-Dist: catboost<1.3,>=1.2; extra == "catboost"
@@ -64,7 +66,7 @@ Requires-Dist: imodels<1.4.0,>=1.3.10; extra == "imodels"
 Provides-Extra: lightgbm
 Requires-Dist: lightgbm<4.6,>=4.0; extra == "lightgbm"
 Provides-Extra: ray
-Requires-Dist: autogluon.core[all]==1.2.1b20250116; extra == "ray"
+Requires-Dist: autogluon.core[all]==1.2.1b20250130; extra == "ray"
 Provides-Extra: skex
 Requires-Dist: scikit-learn-intelex<2025.1,>=2024.0; extra == "skex"
 Provides-Extra: skl2onnx

{autogluon.tabular-1.2.1b20250116.dist-info → autogluon.tabular-1.2.1b20250130.dist-info}/RECORD

@@ -1,6 +1,6 @@
-autogluon.tabular-1.2.1b20250116-py3.8-nspkg.pth,sha256=cQGwpuGPqg1GXscIwt-7PmME1OnSpD-7ixkikJ31WAY,554
+autogluon.tabular-1.2.1b20250130-py3.9-nspkg.pth,sha256=cQGwpuGPqg1GXscIwt-7PmME1OnSpD-7ixkikJ31WAY,554
 autogluon/tabular/__init__.py,sha256=2OXpJCvENRHubBTYNIPpHX93WWuFZzsJBtTZbNVHVas,400
-autogluon/tabular/version.py,sha256=qtf1yrCzRdeJXfqZGb9YWrrjOFhl2omGSnhF5i131QY,91
+autogluon/tabular/version.py,sha256=onZWyUgu5fGg96se5Au1iUkaP0LXAhrwgAuo6bELN9c,91
 autogluon/tabular/configs/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/configs/config_helper.py,sha256=Pb2aW9Z9w77pYKPRVZ3nBzHY3KJaiEJSJ747zZcJIVk,21132
 autogluon/tabular/configs/feature_generator_presets.py,sha256=EV5Ym8VW15q92MwOUpTi7wZFS2QooM51fLg3RdUsn-M,1223
@@ -10,20 +10,20 @@ autogluon/tabular/configs/zeroshot/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQe
 autogluon/tabular/configs/zeroshot/zeroshot_portfolio_2023.py,sha256=oKO_2nEpI_EiLaUGmNN-3kPBIp5ATndbCOaVZ1m0048,29911
 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=KgYEQtbrckWM19MgYIwPSjhe-R-vGgRswNcoLu0wF7M,2552
-autogluon/tabular/experimental/_tabular_regressor.py,sha256=yNJsX2GTmhf8ZTIaHt3GPcjRAdftahp8zM_iY8qkRsA,1957
+autogluon/tabular/experimental/_tabular_classifier.py,sha256=7lGoFdvkHiZS3VpcXo97q4ENV9qyIVDExlWkm0wzL3s,2527
+autogluon/tabular/experimental/_tabular_regressor.py,sha256=EzEDL-19T5QUVNmLkSHNzzGwYrUxyqlNpIDPMgtV6Gg,1932
 autogluon/tabular/learner/__init__.py,sha256=Hhmk5WpKQHohVmI-veOaKMelKJpIdzeXrmw_DPn3DTU,63
 autogluon/tabular/learner/abstract_learner.py,sha256=3myDh867x-EWTPR-O-iw82WGgd5n1NKWf3kaTcYQeh0,54955
 autogluon/tabular/learner/default_learner.py,sha256=cg3K0oA-4ccXWtmGgy6qUJaLldieFwDvnMP_PyE9gdk,24579
 autogluon/tabular/models/__init__.py,sha256=tDVqwVG9q2ctLWRouyXeYs5NiSBnOnwh3anAfZyM3jg,1099
 autogluon/tabular/models/_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autogluon/tabular/models/_utils/rapids_utils.py,sha256=zL1XREBRjFlW592WY5S8_dkQZxCzDW2VDQ-6f9ETCOc,1039
+autogluon/tabular/models/_utils/rapids_utils.py,sha256=gbej9Hjn4alCWZuGN9sOLXMMAyWbgHPThTsp2feS39o,1038
 autogluon/tabular/models/_utils/torch_utils.py,sha256=dxs_KMMAOmNkRNjYf_hrzqaHIfkqn1xoKRKqCFbQ1Rk,537
 autogluon/tabular/models/automm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/automm/automm_model.py,sha256=z16uBs9SAGxuCDbROhTNOTzc4KhCd-ejX31Mt_FHNHY,11159
 autogluon/tabular/models/automm/ft_transformer.py,sha256=OxD5K8xZFxt8PutcGmjaZsAO2QoPpdDG0YoaxnTqrV0,3840
 autogluon/tabular/models/catboost/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autogluon/tabular/models/catboost/callbacks.py,sha256=NwDi82JOFojNS7GV-hf0ZE__KHFFM0TowhNqBjEQV3U,6458
+autogluon/tabular/models/catboost/callbacks.py,sha256=l8x17n_w7oEFs-iDECSdBKZ89yW5g1z-zvj4XLgQPkw,7098
 autogluon/tabular/models/catboost/catboost_model.py,sha256=mURn4GGGXgKovsctXLw3ZPva9xVS_grIm2Lcrj0CdSk,17043
 autogluon/tabular/models/catboost/catboost_softclass_utils.py,sha256=UiW0SUb3hFueW5qYtQn6Sbk7Wg7BWN4jqKWeFtbMvgU,3919
 autogluon/tabular/models/catboost/catboost_utils.py,sha256=YSc94V4DjrwbmkeUM8306zV7z21oq-K-qGCOj0UE_wg,3167
@@ -35,7 +35,7 @@ autogluon/tabular/models/fastainn/callbacks.py,sha256=3WvOEwqd1YAVInooKsFOTzAkCL
 autogluon/tabular/models/fastainn/fastai_helpers.py,sha256=gGYzyrAFl8hi8GnsemZNLGZn5xr7cyJXdFl08PIlza4,1393
 autogluon/tabular/models/fastainn/imports_helper.py,sha256=ICxA8ty47-oZu0Q9AjKCQe8uVi340Iu0NFruxvJPrbA,330
 autogluon/tabular/models/fastainn/quantile_helpers.py,sha256=d89GKvSRBgOy9EqcDI83MK5sqPRxP6JJ3BmPLmKnB0o,1808
-autogluon/tabular/models/fastainn/tabular_nn_fastai.py,sha256=ZdjZLEcHZTZdLpwX8pbVTe1xo5m7oKzUDIXT9yLcTNA,28963
+autogluon/tabular/models/fastainn/tabular_nn_fastai.py,sha256=ZGHpa6FY-3NxDUVQk5fTBeiIt735xEGUbuPgELxij_8,28957
 autogluon/tabular/models/fastainn/hyperparameters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/fastainn/hyperparameters/parameters.py,sha256=DkQwAZZ7CuODKoljr-yrkx-uFxBSPRxkKuvPdwO-UhQ,2069
 autogluon/tabular/models/fastainn/hyperparameters/searchspaces.py,sha256=5qdknZDrHtdPdrhSqjamYQrCxvupXvlN3bVGEPgs48E,1660
@@ -55,7 +55,7 @@ autogluon/tabular/models/knn/knn_utils.py,sha256=XU1cxVXp1BAoQnja2_KmSIn9_q9gZkj
 autogluon/tabular/models/lgb/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/lgb/callbacks.py,sha256=0X42-nAbftKnu_zmFPDf8S3RrUJJjsJ1Qs_TPAJxzjU,11367
 autogluon/tabular/models/lgb/lgb_model.py,sha256=udmA6A-2RhVLRcaJXbjZCWrOhansButwRPRO7u0-BnM,24715
-autogluon/tabular/models/lgb/lgb_utils.py,sha256=Uh-4t3mxhgRSGpCaj-ah7gk6PW7zp-jh8HkskVSYuoA,7388
+autogluon/tabular/models/lgb/lgb_utils.py,sha256=jzTDTzP-z7gcBGZyy1_0YkyTOLbU5DLeRqtil4FCZPI,7382
 autogluon/tabular/models/lgb/hyperparameters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/lgb/hyperparameters/parameters.py,sha256=LLEQ-Ns3HElWBsFJx3ogRV7L6qw_nXlcl7EyO0C0fVQ,1336
 autogluon/tabular/models/lgb/hyperparameters/searchspaces.py,sha256=tvNNR7niWz_B-PndYQXb6vVNABxSfBYRHj6ZVQJ1x2E,1930
@@ -74,11 +74,11 @@ 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/tab_transformer/__init__.py,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
-autogluon/tabular/models/tab_transformer/modified_transformer.py,sha256=QfKHsTY-l2NRwfaiLq1vbUvok1nAOF0m4MQqV_xNXrc,22901
-autogluon/tabular/models/tab_transformer/pretexts.py,sha256=SKcPRA8iowhJP6m7Gkm4jiBqXTkrNnXfp4ZlHyuLx6U,6570
+autogluon/tabular/models/tab_transformer/modified_transformer.py,sha256=3UNhZqKChkWCPtSP4YP23JhtggLqa4mLlP__jvxfBko,22894
+autogluon/tabular/models/tab_transformer/pretexts.py,sha256=UEoDq_8hLKbY7EbE5IyL_gUXpuQ607XTtS-jKqf8j8U,6564
 autogluon/tabular/models/tab_transformer/tab_model_base.py,sha256=4rmY1IrwoFuJejy-9gOoYSz-ar3DvZY8uXyDUBKk7Iw,3615
 autogluon/tabular/models/tab_transformer/tab_transformer.py,sha256=1c1oTJfSsGxQjzZJVN8doqFmYV-Wwwbqcu7RcW77kJk,6991
-autogluon/tabular/models/tab_transformer/tab_transformer_encoder.py,sha256=hu0hrxvQOcGEuRk0w2BZgk-CBCXK4YqzcBH0mHbXah8,24707
+autogluon/tabular/models/tab_transformer/tab_transformer_encoder.py,sha256=v2G1S_MSESzKqtvSfxS5uEse2CWtOn_K2E-uIwuE6zI,24701
 autogluon/tabular/models/tab_transformer/tab_transformer_model.py,sha256=pwqoig6SL3PutYhsB1o3mFyvBubXm91ynPGujp3Ab3g,22702
 autogluon/tabular/models/tab_transformer/utils.py,sha256=rrNk0X6Y0vzt7ivylRmTf0zjDx1DIF-5_ibf6B9Taz8,4554
 autogluon/tabular/models/tab_transformer/hyperparameters/__init__.py,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
@@ -121,7 +121,7 @@ autogluon/tabular/models/tabular_nn/hyperparameters/parameters.py,sha256=Z3t_U1f
 autogluon/tabular/models/tabular_nn/hyperparameters/searchspaces.py,sha256=pT9cJ3MaWPnaQwAf47Yz6f0-L9qDBknahERbggAp52U,2810
 autogluon/tabular/models/tabular_nn/torch/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/tabular_nn/torch/tabular_nn_torch.py,sha256=1ZJHM52wMZHSbXty19ykKA5tNql6ISO7nuFmAAuR-uU,42324
-autogluon/tabular/models/tabular_nn/torch/tabular_torch_dataset.py,sha256=oelC0uA9KNVtNKXU5jTywg-OfIF-5AguAXFYSKwN3zU,13499
+autogluon/tabular/models/tabular_nn/torch/tabular_torch_dataset.py,sha256=RdnQGZSrvY1iuJB4JTANniH3Dorw-DP0Em_JK3_h7RM,13497
 autogluon/tabular/models/tabular_nn/torch/torch_network_modules.py,sha256=Qc3PwXTD8A7PgXi6EGuaBCrN3jsFAXDLCW7i6tE5wYI,11338
 autogluon/tabular/models/tabular_nn/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/tabular_nn/utils/categorical_encoders.py,sha256=uLQaHkudg8AS2jFwNv4FyQ9KzO_F-dj6hIJe8R2t4aw,34213
@@ -142,21 +142,21 @@ autogluon/tabular/models/xgboost/hyperparameters/searchspaces.py,sha256=lFwI34pc
 autogluon/tabular/models/xt/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/xt/xt_model.py,sha256=gimCfMeTmhAJTy4ekCI_dbK1FluJ7EQMoHt0pGMvVgs,759
 autogluon/tabular/predictor/__init__.py,sha256=zCMgjxQlWpDWnr1l1xjBCiK3rWC3N3RoD8UXBnazT74,107
-autogluon/tabular/predictor/_deprecated_methods.py,sha256=N0QqI9wElXdSeQ9idnnJpoH2xkmv51j_vG7kDrP5QxU,3507
+autogluon/tabular/predictor/_deprecated_methods.py,sha256=cBZjVYxY7blOj2eH1y9JsdW0n-AZtnsK4B6dzUqrTAw,3476
 autogluon/tabular/predictor/interpretable_predictor.py,sha256=5UeKgnMFsfY65tiO3kxfHBPr03lyswLrgdtjPhI0Y7Q,6934
-autogluon/tabular/predictor/predictor.py,sha256=-0d5g_ADcSCwJKS6uJVif9iJD3Sq6FgcVFPrPRZ-l1Q,352445
+autogluon/tabular/predictor/predictor.py,sha256=uFyhlsJ2T8kzeVUOVAfRsR_DoyumEVgBYNp44hGHuPI,352421
 autogluon/tabular/trainer/__init__.py,sha256=PW_PGL-tWoQzx3ES2S53bQEZOtsRWTYiM9QdOqsk0dI,38
-autogluon/tabular/trainer/auto_trainer.py,sha256=ieu7Zofl93tnQC9sIcsvS9pJS__KVIsE7Nn-EIWRvT0,8718
+autogluon/tabular/trainer/auto_trainer.py,sha256=3cRi079OoFbXYJfnU2evE2502E-Zz-jXuAKwwOHo3w4,8702
 autogluon/tabular/trainer/model_presets/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/trainer/model_presets/presets.py,sha256=1E-Z1FxUpyydaoEdxcTCg7bvSwVEuwImhH2-VlcRdaQ,19217
 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-1.2.1b20250116.dist-info/LICENSE,sha256=CeipvOyAZxBGUsFoaFqwkx54aPnIKEtm9a5u2uXxEws,10142
-autogluon.tabular-1.2.1b20250116.dist-info/METADATA,sha256=QSKAy7j9NWzgoqH_awYRFKYq2fif9r603N26Xl6eweM,14315
-autogluon.tabular-1.2.1b20250116.dist-info/NOTICE,sha256=7nPQuj8Kp-uXsU0S5so3-2dNU5EctS5hDXvvzzehd7E,114
-autogluon.tabular-1.2.1b20250116.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-autogluon.tabular-1.2.1b20250116.dist-info/namespace_packages.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
-autogluon.tabular-1.2.1b20250116.dist-info/top_level.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
-autogluon.tabular-1.2.1b20250116.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
-autogluon.tabular-1.2.1b20250116.dist-info/RECORD,,
+autogluon.tabular-1.2.1b20250130.dist-info/LICENSE,sha256=CeipvOyAZxBGUsFoaFqwkx54aPnIKEtm9a5u2uXxEws,10142
+autogluon.tabular-1.2.1b20250130.dist-info/METADATA,sha256=PhsrAew0RTxS1U1QWqMVQIYDsmegW8Z8MQTaL6GgZEI,14364
+autogluon.tabular-1.2.1b20250130.dist-info/NOTICE,sha256=7nPQuj8Kp-uXsU0S5so3-2dNU5EctS5hDXvvzzehd7E,114
+autogluon.tabular-1.2.1b20250130.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+autogluon.tabular-1.2.1b20250130.dist-info/namespace_packages.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
+autogluon.tabular-1.2.1b20250130.dist-info/top_level.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
+autogluon.tabular-1.2.1b20250130.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
+autogluon.tabular-1.2.1b20250130.dist-info/RECORD,,