autogluon.tabular 1.2.1b20250425__py3-none-any.whl → 1.2.1b20250426__py3-none-any.whl

autogluon/tabular/models/catboost/catboost_model.py

@@ -95,17 +95,26 @@ class CatBoostModel(AbstractModel):
             Scales roughly by 5080*num_features*2^depth bytes
             For 10000 features and 6 depth, the histogram would be 3.2 GB.
         """
+        if hyperparameters is None:
+            hyperparameters = {}
         num_classes = num_classes if num_classes else 1  # self.num_classes could be None after initialization if it's a regression problem
         data_mem_usage = get_approximate_df_mem_usage(X).sum()
         data_mem_usage_bytes = data_mem_usage * 5 + data_mem_usage / 4 * num_classes  # TODO: Extremely crude approximation, can be vastly improved
 
         border_count = hyperparameters.get("border_count", 254)
         depth = hyperparameters.get("depth", 6)
+
+        # if depth < 7, treat it as 1 step larger for histogram size estimate
+        #  this fixes cases where otherwise histogram size appears to be off by around a factor of 2 for depth=6
+        histogram_effective_depth = max(min(depth+1, 7), depth)
+
         # Formula based on manual testing, aligns with LightGBM histogram sizes
-        histogram_mem_usage_bytes = 20 * math.pow(2, depth) * len(X.columns) * border_count
+        histogram_mem_usage_bytes = 24 * math.pow(2, histogram_effective_depth) * len(X.columns) * border_count
         histogram_mem_usage_bytes *= 1.2  # Add a 20% buffer
 
-        approx_mem_size_req = data_mem_usage_bytes + histogram_mem_usage_bytes
+        baseline_memory_bytes = 4e8  # 400 MB baseline memory
+
+        approx_mem_size_req = data_mem_usage_bytes + histogram_mem_usage_bytes + baseline_memory_bytes
         return approx_mem_size_req
 
     # TODO: Use Pool in preprocess, optimize bagging to do Pool.split() to avoid re-computing pool for each fold! Requires stateful + y

autogluon/tabular/models/lgb/lgb_model.py

@@ -3,7 +3,6 @@ from __future__ import annotations
 import gc
 import logging
 import os
-import random
 import re
 import time
 import warnings
@@ -99,6 +98,8 @@ class LGBModel(AbstractModel):
             Scales roughly by 5100*num_features*num_leaves bytes
             For 10000 features and 128 num_leaves, the histogram would be 6.5 GB.
         """
+        if hyperparameters is None:
+            hyperparameters = {}
         num_classes = num_classes if num_classes else 1  # num_classes could be None after initialization if it's a regression problem
         data_mem_usage = get_approximate_df_mem_usage(X).sum()
         data_mem_usage_bytes = data_mem_usage * 5 + data_mem_usage / 4 * num_classes  # TODO: Extremely crude approximation, can be vastly improved
@@ -275,8 +276,6 @@ class LGBModel(AbstractModel):
             train_params["params"]["quantile_levels"] = self.quantile_levels
         if seed_val is not None:
             train_params["params"]["seed"] = seed_val
-            random.seed(seed_val)
-            np.random.seed(seed_val)
 
         # Train LightGBM model:
         # Note that self.model contains a <class 'lightgbm.basic.Booster'> not a LightBGMClassifier or LightGBMRegressor object

autogluon/tabular/models/rf/rf_model.py

@@ -147,6 +147,8 @@ class RFModel(AbstractModel):
         num_classes: int = 1,
         **kwargs,
     ) -> int:
+        if hyperparameters is None:
+            hyperparameters = {}
         n_estimators_final = hyperparameters.get("n_estimators", 300)
         if isinstance(n_estimators_final, int):
             n_estimators_minimum = min(40, n_estimators_final)

autogluon/tabular/models/xgboost/callbacks.py

@@ -147,7 +147,7 @@ class EarlyStoppingCustom(EarlyStopping):
 
         model_size_memory_ratio = estimated_model_size_mb / available_mb
 
-        if (model_size_memory_ratio > 1.0) or (available_mb < 512):
+        if (model_size_memory_ratio > 0.75) or (available_mb < 512):
             logger.warning("Warning: Large XGB model size may cause OOM error if training continues")
             logger.warning(f"Available Memory: {available_mb} MB")
             logger.warning(f"Estimated XGB model size: {estimated_model_size_mb} MB")

autogluon/tabular/models/xgboost/xgboost_model.py

@@ -256,12 +256,25 @@ class XGBoostModel(AbstractModel):
         num_classes: int = 1,
         **kwargs,
     ) -> int:
+        if hyperparameters is None:
+            hyperparameters = {}
         num_classes = num_classes if num_classes else 1  # self.num_classes could be None after initialization if it's a regression problem
         data_mem_usage = get_approximate_df_mem_usage(X).sum()
         data_mem_usage_bytes = data_mem_usage * 7 + data_mem_usage / 4 * num_classes  # TODO: Extremely crude approximation, can be vastly improved
 
         max_bin = hyperparameters.get("max_bin", 256)
         max_depth = hyperparameters.get("max_depth", 6)
+        max_leaves = hyperparameters.get("max_leaves", 0)
+        if max_leaves is None:
+            max_leaves = 0
+
+        if max_depth > 12 or max_depth == 0:  # 0 = uncapped
+            max_depth = 12  # Try our best if the value is very large, only treat it as 12.
+
+        if max_leaves != 0:  # if capped max_leaves
+            # make the effective max_depth for calculations be the lesser of the two constraints
+            max_depth = min(max_depth, math.ceil(math.log2(max_leaves)))
+
         # Formula based on manual testing, aligns with LightGBM histogram sizes
         #  This approximation is less accurate than it is for LightGBM and CatBoost.
         #  Note that max_depth didn't appear to reduce memory usage below 6, and it was unclear if it increased memory usage above 6.
@@ -274,7 +287,12 @@ class XGBoostModel(AbstractModel):
         histogram_mem_usage_bytes = 20 * depth_modifier * len(X.columns) * max_bin
         histogram_mem_usage_bytes *= 1.2  # Add a 20% buffer
 
-        approx_mem_size_req = data_mem_usage_bytes + histogram_mem_usage_bytes
+        mem_size_per_estimator = num_classes * max_depth * 500  # very rough estimate
+        n_estimators = hyperparameters.get("n_estimators", 10000)
+        n_estimators_min = min(n_estimators, 1000)
+        mem_size_estimators = n_estimators_min * mem_size_per_estimator  # memory estimate after fitting up to 1000 estimators
+
+        approx_mem_size_req = data_mem_usage_bytes + histogram_mem_usage_bytes + mem_size_estimators
         return approx_mem_size_req
 
     def _validate_fit_memory_usage(self, mem_error_threshold: float = 1.0, mem_warning_threshold: float = 0.75, mem_size_threshold: int = 1e9, **kwargs):

autogluon/tabular/version.py

@@ -1,4 +1,4 @@
 """This is the autogluon version file."""
 
-__version__ = "1.2.1b20250425"
+__version__ = "1.2.1b20250426"
 __lite__ = False

{autogluon.tabular-1.2.1b20250425.dist-info → autogluon.tabular-1.2.1b20250426.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: autogluon.tabular
-Version: 1.2.1b20250425
+Version: 1.2.1b20250426
 Summary: Fast and Accurate ML in 3 Lines of Code
 Home-page: https://github.com/autogluon/autogluon
 Author: AutoGluon Community
@@ -41,19 +41,19 @@ Requires-Dist: scipy<1.16,>=1.5.4
 Requires-Dist: pandas<2.3.0,>=2.0.0
 Requires-Dist: scikit-learn<1.7.0,>=1.4.0
 Requires-Dist: networkx<4,>=3.0
-Requires-Dist: autogluon.core==1.2.1b20250425
-Requires-Dist: autogluon.features==1.2.1b20250425
+Requires-Dist: autogluon.core==1.2.1b20250426
+Requires-Dist: autogluon.features==1.2.1b20250426
 Provides-Extra: all
-Requires-Dist: catboost<1.3,>=1.2; extra == "all"
-Requires-Dist: autogluon.core[all]==1.2.1b20250425; extra == "all"
-Requires-Dist: torch<2.7,>=2.2; extra == "all"
-Requires-Dist: xgboost<3.1,>=2.0; extra == "all"
 Requires-Dist: einops<0.9,>=0.7; extra == "all"
-Requires-Dist: huggingface-hub[torch]; extra == "all"
+Requires-Dist: torch<2.7,>=2.2; extra == "all"
 Requires-Dist: lightgbm<4.7,>=4.0; extra == "all"
+Requires-Dist: huggingface-hub[torch]; extra == "all"
 Requires-Dist: numpy<2.0.0,>=1.25; extra == "all"
 Requires-Dist: fastai<2.9,>=2.3.1; extra == "all"
 Requires-Dist: spacy<3.8; extra == "all"
+Requires-Dist: xgboost<3.1,>=2.0; extra == "all"
+Requires-Dist: autogluon.core[all]==1.2.1b20250426; extra == "all"
+Requires-Dist: catboost<1.3,>=1.2; extra == "all"
 Provides-Extra: catboost
 Requires-Dist: numpy<2.0.0,>=1.25; extra == "catboost"
 Requires-Dist: catboost<1.3,>=1.2; extra == "catboost"
@@ -66,7 +66,7 @@ Requires-Dist: imodels<2.1.0,>=1.3.10; extra == "imodels"
 Provides-Extra: lightgbm
 Requires-Dist: lightgbm<4.7,>=4.0; extra == "lightgbm"
 Provides-Extra: ray
-Requires-Dist: autogluon.core[all]==1.2.1b20250425; extra == "ray"
+Requires-Dist: autogluon.core[all]==1.2.1b20250426; extra == "ray"
 Provides-Extra: skex
 Requires-Dist: scikit-learn-intelex<2025.5,>=2024.0; extra == "skex"
 Provides-Extra: skl2onnx

{autogluon.tabular-1.2.1b20250425.dist-info → autogluon.tabular-1.2.1b20250426.dist-info}/RECORD

@@ -1,6 +1,6 @@
-autogluon.tabular-1.2.1b20250425-py3.9-nspkg.pth,sha256=cQGwpuGPqg1GXscIwt-7PmME1OnSpD-7ixkikJ31WAY,554
+autogluon.tabular-1.2.1b20250426-py3.9-nspkg.pth,sha256=cQGwpuGPqg1GXscIwt-7PmME1OnSpD-7ixkikJ31WAY,554
 autogluon/tabular/__init__.py,sha256=2OXpJCvENRHubBTYNIPpHX93WWuFZzsJBtTZbNVHVas,400
-autogluon/tabular/version.py,sha256=zTyiK3ClhRyp8BZjIMLB2n9YsA7GaR9qnPDLREdyRQs,91
+autogluon/tabular/version.py,sha256=OsP4ql15uiax-GUPsVJ1sGZMaTOWMBXWWMnLOe5tRrY,91
 autogluon/tabular/configs/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/configs/config_helper.py,sha256=wLgu94NjV-l2fwZacpKqjPfvk8E_RwAl_L1hfH5xO8E,21085
 autogluon/tabular/configs/feature_generator_presets.py,sha256=EV5Ym8VW15q92MwOUpTi7wZFS2QooM51fLg3RdUsn-M,1223
@@ -25,7 +25,7 @@ autogluon/tabular/models/automm/automm_model.py,sha256=GvrMBC8Z-zobalmSzX1iDHTYM
 autogluon/tabular/models/automm/ft_transformer.py,sha256=yZ9-TTA4GbtutHhz0Djkrl-rIFNxc7A2LBOFOXYOxVY,3886
 autogluon/tabular/models/catboost/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/catboost/callbacks.py,sha256=l8x17n_w7oEFs-iDECSdBKZ89yW5g1z-zvj4XLgQPkw,7098
-autogluon/tabular/models/catboost/catboost_model.py,sha256=ApltUyLY0BUuS-Rb8z2_OgkAniXrm3Zlpd9FDZkRMjM,17416
+autogluon/tabular/models/catboost/catboost_model.py,sha256=Dv62XDuJ_sFvU95xeY1gHuV_F2RoYDGaFa6-4XNwnnU,17849
 autogluon/tabular/models/catboost/catboost_softclass_utils.py,sha256=UiW0SUb3hFueW5qYtQn6Sbk7Wg7BWN4jqKWeFtbMvgU,3919
 autogluon/tabular/models/catboost/catboost_utils.py,sha256=YSc94V4DjrwbmkeUM8306zV7z21oq-K-qGCOj0UE_wg,3167
 autogluon/tabular/models/catboost/hyperparameters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -55,7 +55,7 @@ autogluon/tabular/models/knn/knn_rapids_model.py,sha256=0FFApNZFH8nyrDqlBSUV7jO-
 autogluon/tabular/models/knn/knn_utils.py,sha256=XU1cxVXp1BAoQnja2_KmSIn9_q9gZkjAya7-9b0uStk,7455
 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=3hIk24x3HZErGKW9VfgN7-qu92-fovjATXqhxh81EPo,25053
+autogluon/tabular/models/lgb/lgb_model.py,sha256=jBoku48tvxqWToCc0qUZWRhTWPUHxcDm_8ku435_eSg,25037
 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
@@ -68,7 +68,7 @@ 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/rf/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autogluon/tabular/models/rf/rf_model.py,sha256=MwhTUk9PknLgZWyT5bz_7DtjEW8ADU5wiBvlOjzVLkA,21529
+autogluon/tabular/models/rf/rf_model.py,sha256=2iG45F0k17No7ycrfVMWBfU_WqJwC4MTehaFGsguLzE,21598
 autogluon/tabular/models/rf/rf_quantile.py,sha256=2S8FE8po9lMnZaeKuVkzOUFOcdil46ZbFqm49OuvNZY,36460
 autogluon/tabular/models/rf/rf_rapids_model.py,sha256=3s-8M11dzCl_2Lu5iB3H8YjHLgyP_SElrm_4w_HfmqY,2028
 autogluon/tabular/models/rf/compilers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -120,8 +120,8 @@ autogluon/tabular/models/tabular_nn/utils/nn_architecture_utils.py,sha256=tttzR5
 autogluon/tabular/models/text_prediction/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/text_prediction/text_prediction_v1_model.py,sha256=PBN7F98qgEAO6U76rV_hxZfAmKr_XpVKjElOdBvfX8c,1090
 autogluon/tabular/models/xgboost/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autogluon/tabular/models/xgboost/callbacks.py,sha256=uynimXya07XQMBkDvec-7mXK6OfMGP6M8MiVYu8OVRI,7008
-autogluon/tabular/models/xgboost/xgboost_model.py,sha256=tTSnTzEot2JB0qEvhki4h3RdaLjEpfMs-jWKsxlJWO4,14304
+autogluon/tabular/models/xgboost/callbacks.py,sha256=PuRQUg3AEjgvFa-dpstRFoEVM9jHDe5W4XYSdDPRqoE,7009
+autogluon/tabular/models/xgboost/xgboost_model.py,sha256=_HFwfEbAg0CllEUEk9HVLsXXVBcuC_qsGd7z7arJYPs,15220
 autogluon/tabular/models/xgboost/xgboost_utils.py,sha256=FVqZ8h4JAe_pifSvNx83cLZHwsuzTXylrrcan07AoNo,5757
 autogluon/tabular/models/xgboost/hyperparameters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autogluon/tabular/models/xgboost/hyperparameters/parameters.py,sha256=ay6bVVpiPzftbtz6TTS76w7j4vjDjzHFpuf2Bjf6Zu4,1673
@@ -146,11 +146,11 @@ autogluon/tabular/trainer/model_presets/presets.py,sha256=IMBRSBc-djd35gkb2rtXrW
 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-1.2.1b20250425.dist-info/LICENSE,sha256=CeipvOyAZxBGUsFoaFqwkx54aPnIKEtm9a5u2uXxEws,10142
-autogluon.tabular-1.2.1b20250425.dist-info/METADATA,sha256=WXVB7wtSVG30KkMeusKLaBZaU7FWDJrrXbmC2rCH6Xs,14069
-autogluon.tabular-1.2.1b20250425.dist-info/NOTICE,sha256=7nPQuj8Kp-uXsU0S5so3-2dNU5EctS5hDXvvzzehd7E,114
-autogluon.tabular-1.2.1b20250425.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-autogluon.tabular-1.2.1b20250425.dist-info/namespace_packages.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
-autogluon.tabular-1.2.1b20250425.dist-info/top_level.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
-autogluon.tabular-1.2.1b20250425.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
-autogluon.tabular-1.2.1b20250425.dist-info/RECORD,,
+autogluon.tabular-1.2.1b20250426.dist-info/LICENSE,sha256=CeipvOyAZxBGUsFoaFqwkx54aPnIKEtm9a5u2uXxEws,10142
+autogluon.tabular-1.2.1b20250426.dist-info/METADATA,sha256=g4AxiS8dTRMHEO6FuxALlnv6-njEhYubb6rBmohshOo,14069
+autogluon.tabular-1.2.1b20250426.dist-info/NOTICE,sha256=7nPQuj8Kp-uXsU0S5so3-2dNU5EctS5hDXvvzzehd7E,114
+autogluon.tabular-1.2.1b20250426.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+autogluon.tabular-1.2.1b20250426.dist-info/namespace_packages.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
+autogluon.tabular-1.2.1b20250426.dist-info/top_level.txt,sha256=giERA4R78OkJf2ijn5slgjURlhRPzfLr7waIcGkzYAo,10
+autogluon.tabular-1.2.1b20250426.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
+autogluon.tabular-1.2.1b20250426.dist-info/RECORD,,