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

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.

This version of autogluon.tabular might be problematic. Click here for more details.

Files changed (135) hide show
  1. autogluon/tabular/__init__.py +1 -0
  2. autogluon/tabular/configs/config_helper.py +18 -6
  3. autogluon/tabular/configs/feature_generator_presets.py +3 -1
  4. autogluon/tabular/configs/hyperparameter_configs.py +42 -9
  5. autogluon/tabular/configs/presets_configs.py +38 -14
  6. autogluon/tabular/configs/zeroshot/zeroshot_portfolio_2023.py +84 -14
  7. autogluon/tabular/configs/zeroshot/zeroshot_portfolio_2025.py +48 -48
  8. autogluon/tabular/configs/zeroshot/zeroshot_portfolio_cpu_2025_12_18.py +774 -1
  9. autogluon/tabular/configs/zeroshot/zeroshot_portfolio_gpu_2025_12_18.py +421 -1
  10. autogluon/tabular/experimental/_scikit_mixin.py +6 -2
  11. autogluon/tabular/experimental/_tabular_classifier.py +3 -1
  12. autogluon/tabular/experimental/_tabular_regressor.py +3 -1
  13. autogluon/tabular/experimental/plot_leaderboard.py +73 -19
  14. autogluon/tabular/learner/abstract_learner.py +160 -42
  15. autogluon/tabular/learner/default_learner.py +78 -22
  16. autogluon/tabular/models/__init__.py +2 -2
  17. autogluon/tabular/models/_utils/rapids_utils.py +3 -1
  18. autogluon/tabular/models/abstract/abstract_torch_model.py +2 -0
  19. autogluon/tabular/models/automm/automm_model.py +12 -3
  20. autogluon/tabular/models/automm/ft_transformer.py +5 -1
  21. autogluon/tabular/models/catboost/callbacks.py +2 -2
  22. autogluon/tabular/models/catboost/catboost_model.py +93 -29
  23. autogluon/tabular/models/catboost/catboost_softclass_utils.py +4 -1
  24. autogluon/tabular/models/catboost/catboost_utils.py +3 -1
  25. autogluon/tabular/models/ebm/ebm_model.py +8 -13
  26. autogluon/tabular/models/ebm/hyperparameters/parameters.py +1 -0
  27. autogluon/tabular/models/ebm/hyperparameters/searchspaces.py +1 -0
  28. autogluon/tabular/models/fastainn/callbacks.py +20 -3
  29. autogluon/tabular/models/fastainn/hyperparameters/searchspaces.py +11 -1
  30. autogluon/tabular/models/fastainn/quantile_helpers.py +10 -2
  31. autogluon/tabular/models/fastainn/tabular_nn_fastai.py +65 -18
  32. autogluon/tabular/models/fasttext/fasttext_model.py +3 -1
  33. autogluon/tabular/models/image_prediction/image_predictor.py +7 -2
  34. autogluon/tabular/models/knn/knn_model.py +41 -8
  35. autogluon/tabular/models/lgb/callbacks.py +32 -9
  36. autogluon/tabular/models/lgb/hyperparameters/searchspaces.py +3 -1
  37. autogluon/tabular/models/lgb/lgb_model.py +150 -34
  38. autogluon/tabular/models/lgb/lgb_utils.py +12 -4
  39. autogluon/tabular/models/lr/hyperparameters/searchspaces.py +5 -1
  40. autogluon/tabular/models/lr/lr_model.py +40 -10
  41. autogluon/tabular/models/lr/lr_rapids_model.py +22 -13
  42. autogluon/tabular/models/mitra/_internal/__init__.py +1 -1
  43. autogluon/tabular/models/mitra/_internal/config/__init__.py +1 -1
  44. autogluon/tabular/models/mitra/_internal/config/config_pretrain.py +36 -40
  45. autogluon/tabular/models/mitra/_internal/config/config_run.py +2 -14
  46. autogluon/tabular/models/mitra/_internal/config/enums.py +27 -26
  47. autogluon/tabular/models/mitra/_internal/core/__init__.py +1 -1
  48. autogluon/tabular/models/mitra/_internal/core/callbacks.py +14 -21
  49. autogluon/tabular/models/mitra/_internal/core/get_loss.py +10 -12
  50. autogluon/tabular/models/mitra/_internal/core/get_optimizer.py +17 -32
  51. autogluon/tabular/models/mitra/_internal/core/get_scheduler.py +12 -27
  52. autogluon/tabular/models/mitra/_internal/core/prediction_metrics.py +16 -21
  53. autogluon/tabular/models/mitra/_internal/core/trainer_finetune.py +130 -111
  54. autogluon/tabular/models/mitra/_internal/data/__init__.py +1 -1
  55. autogluon/tabular/models/mitra/_internal/data/collator.py +30 -26
  56. autogluon/tabular/models/mitra/_internal/data/dataset_finetune.py +18 -26
  57. autogluon/tabular/models/mitra/_internal/data/dataset_split.py +10 -7
  58. autogluon/tabular/models/mitra/_internal/data/preprocessor.py +70 -100
  59. autogluon/tabular/models/mitra/_internal/models/__init__.py +1 -1
  60. autogluon/tabular/models/mitra/_internal/models/base.py +7 -10
  61. autogluon/tabular/models/mitra/_internal/models/embedding.py +46 -56
  62. autogluon/tabular/models/mitra/_internal/models/tab2d.py +140 -120
  63. autogluon/tabular/models/mitra/_internal/utils/__init__.py +1 -1
  64. autogluon/tabular/models/mitra/_internal/utils/set_seed.py +3 -1
  65. autogluon/tabular/models/mitra/mitra_model.py +16 -11
  66. autogluon/tabular/models/mitra/sklearn_interface.py +178 -162
  67. autogluon/tabular/models/realmlp/realmlp_model.py +28 -15
  68. autogluon/tabular/models/rf/compilers/onnx.py +1 -1
  69. autogluon/tabular/models/rf/rf_model.py +45 -12
  70. autogluon/tabular/models/rf/rf_quantile.py +4 -2
  71. autogluon/tabular/models/tabdpt/tabdpt_model.py +8 -17
  72. autogluon/tabular/models/tabicl/tabicl_model.py +8 -1
  73. autogluon/tabular/models/tabm/_tabm_internal.py +6 -4
  74. autogluon/tabular/models/tabm/rtdl_num_embeddings.py +80 -127
  75. autogluon/tabular/models/tabm/tabm_model.py +8 -4
  76. autogluon/tabular/models/tabm/tabm_reference.py +53 -85
  77. autogluon/tabular/models/tabpfnmix/_internal/core/callbacks.py +7 -16
  78. autogluon/tabular/models/tabpfnmix/_internal/core/collator.py +16 -24
  79. autogluon/tabular/models/tabpfnmix/_internal/core/dataset_split.py +5 -7
  80. autogluon/tabular/models/tabpfnmix/_internal/core/enums.py +0 -2
  81. autogluon/tabular/models/tabpfnmix/_internal/core/get_loss.py +0 -1
  82. autogluon/tabular/models/tabpfnmix/_internal/core/get_optimizer.py +7 -18
  83. autogluon/tabular/models/tabpfnmix/_internal/core/get_scheduler.py +3 -14
  84. autogluon/tabular/models/tabpfnmix/_internal/core/trainer_finetune.py +79 -64
  85. autogluon/tabular/models/tabpfnmix/_internal/core/y_transformer.py +3 -5
  86. autogluon/tabular/models/tabpfnmix/_internal/data/dataset_finetune.py +17 -30
  87. autogluon/tabular/models/tabpfnmix/_internal/data/preprocessor.py +15 -35
  88. autogluon/tabular/models/tabpfnmix/_internal/models/foundation/embedding.py +21 -38
  89. autogluon/tabular/models/tabpfnmix/_internal/models/foundation/foundation_transformer.py +33 -51
  90. autogluon/tabular/models/tabpfnmix/_internal/results/prediction_metrics.py +4 -4
  91. autogluon/tabular/models/tabpfnmix/_internal/tabpfnmix_classifier.py +32 -12
  92. autogluon/tabular/models/tabpfnmix/_internal/tabpfnmix_regressor.py +32 -13
  93. autogluon/tabular/models/tabpfnmix/tabpfnmix_model.py +55 -19
  94. autogluon/tabular/models/tabpfnv2/tabpfnv2_5_model.py +21 -48
  95. autogluon/tabular/models/tabprep/prep_mixin.py +34 -26
  96. autogluon/tabular/models/tabular_nn/compilers/onnx.py +36 -8
  97. autogluon/tabular/models/tabular_nn/torch/tabular_nn_torch.py +130 -36
  98. autogluon/tabular/models/tabular_nn/torch/tabular_torch_dataset.py +8 -4
  99. autogluon/tabular/models/tabular_nn/torch/torch_network_modules.py +26 -5
  100. autogluon/tabular/models/tabular_nn/utils/categorical_encoders.py +41 -24
  101. autogluon/tabular/models/tabular_nn/utils/data_preprocessor.py +33 -8
  102. autogluon/tabular/models/tabular_nn/utils/nn_architecture_utils.py +21 -6
  103. autogluon/tabular/models/xgboost/callbacks.py +9 -3
  104. autogluon/tabular/models/xgboost/xgboost_model.py +59 -11
  105. autogluon/tabular/models/xt/xt_model.py +1 -0
  106. autogluon/tabular/predictor/interpretable_predictor.py +3 -1
  107. autogluon/tabular/predictor/predictor.py +409 -128
  108. autogluon/tabular/registry/__init__.py +1 -1
  109. autogluon/tabular/registry/_ag_model_registry.py +4 -5
  110. autogluon/tabular/registry/_model_registry.py +1 -0
  111. autogluon/tabular/testing/fit_helper.py +55 -15
  112. autogluon/tabular/testing/generate_datasets.py +1 -1
  113. autogluon/tabular/testing/model_fit_helper.py +10 -4
  114. autogluon/tabular/trainer/abstract_trainer.py +644 -230
  115. autogluon/tabular/trainer/auto_trainer.py +19 -8
  116. autogluon/tabular/trainer/model_presets/presets.py +33 -9
  117. autogluon/tabular/trainer/model_presets/presets_distill.py +16 -2
  118. autogluon/tabular/version.py +1 -1
  119. {autogluon_tabular-1.5.0b20251228.dist-info → autogluon_tabular-1.5.1b20260116.dist-info}/METADATA +26 -26
  120. {autogluon_tabular-1.5.0b20251228.dist-info → autogluon_tabular-1.5.1b20260116.dist-info}/RECORD +127 -135
  121. autogluon/tabular/models/tabpfnv2/rfpfn/__init__.py +0 -20
  122. autogluon/tabular/models/tabpfnv2/rfpfn/configs.py +0 -40
  123. autogluon/tabular/models/tabpfnv2/rfpfn/scoring_utils.py +0 -201
  124. autogluon/tabular/models/tabpfnv2/rfpfn/sklearn_based_decision_tree_tabpfn.py +0 -1464
  125. autogluon/tabular/models/tabpfnv2/rfpfn/sklearn_based_random_forest_tabpfn.py +0 -747
  126. autogluon/tabular/models/tabpfnv2/rfpfn/sklearn_compat.py +0 -863
  127. autogluon/tabular/models/tabpfnv2/rfpfn/utils.py +0 -106
  128. autogluon/tabular/models/tabpfnv2/tabpfnv2_model.py +0 -466
  129. /autogluon.tabular-1.5.0b20251228-py3.11-nspkg.pth → /autogluon.tabular-1.5.1b20260116-py3.11-nspkg.pth +0 -0
  130. {autogluon_tabular-1.5.0b20251228.dist-info → autogluon_tabular-1.5.1b20260116.dist-info}/WHEEL +0 -0
  131. {autogluon_tabular-1.5.0b20251228.dist-info → autogluon_tabular-1.5.1b20260116.dist-info}/licenses/LICENSE +0 -0
  132. {autogluon_tabular-1.5.0b20251228.dist-info → autogluon_tabular-1.5.1b20260116.dist-info}/licenses/NOTICE +0 -0
  133. {autogluon_tabular-1.5.0b20251228.dist-info → autogluon_tabular-1.5.1b20260116.dist-info}/namespace_packages.txt +0 -0
  134. {autogluon_tabular-1.5.0b20251228.dist-info → autogluon_tabular-1.5.1b20260116.dist-info}/top_level.txt +0 -0
  135. {autogluon_tabular-1.5.0b20251228.dist-info → autogluon_tabular-1.5.1b20260116.dist-info}/zip-safe +0 -0
autogluon/tabular/configs/zeroshot/zeroshot_portfolio_2025.py CHANGED
@@ -2,12 +2,12 @@
2
2
  hyperparameter_portfolio_zeroshot_2025_small = {
3
3
  "REALTABPFN-V2": [
4
4
  {
5
- "ag_args": {'priority': -1},
5
+ "ag_args": {"priority": -1},
6
6
  },
7
7
  ],
8
8
  "GBM": [
9
9
  {
10
- "ag_args": {'name_suffix': '_r33', 'priority': -2},
10
+ "ag_args": {"name_suffix": "_r33", "priority": -2},
11
11
  "bagging_fraction": 0.9625293420216,
12
12
  "bagging_freq": 1,
13
13
  "cat_l2": 0.1236875455555,
@@ -23,7 +23,7 @@ hyperparameter_portfolio_zeroshot_2025_small = {
23
23
  "num_leaves": 68,
24
24
  },
25
25
  {
26
- "ag_args": {'name_suffix': '_r21', 'priority': -16},
26
+ "ag_args": {"name_suffix": "_r21", "priority": -16},
27
27
  "bagging_fraction": 0.7218730663234,
28
28
  "bagging_freq": 1,
29
29
  "cat_l2": 0.0296205152578,
@@ -39,7 +39,7 @@ hyperparameter_portfolio_zeroshot_2025_small = {
39
39
  "num_leaves": 30,
40
40
  },
41
41
  {
42
- "ag_args": {'name_suffix': '_r11', 'priority': -19},
42
+ "ag_args": {"name_suffix": "_r11", "priority": -19},
43
43
  "bagging_fraction": 0.775784726514,
44
44
  "bagging_freq": 1,
45
45
  "cat_l2": 0.3888471449178,
@@ -57,15 +57,15 @@ hyperparameter_portfolio_zeroshot_2025_small = {
57
57
  ],
58
58
  "CAT": [
59
59
  {
60
- "ag_args": {'priority': -5},
60
+ "ag_args": {"priority": -5},
61
61
  },
62
62
  {
63
- "ag_args": {'name_suffix': '_r51', 'priority': -10},
64
- "boosting_type": 'Plain',
65
- "bootstrap_type": 'Bernoulli',
63
+ "ag_args": {"name_suffix": "_r51", "priority": -10},
64
+ "boosting_type": "Plain",
65
+ "bootstrap_type": "Bernoulli",
66
66
  "colsample_bylevel": 0.8771035272558,
67
67
  "depth": 7,
68
- "grow_policy": 'SymmetricTree',
68
+ "grow_policy": "SymmetricTree",
69
69
  "l2_leaf_reg": 2.0107286863021,
70
70
  "leaf_estimation_iterations": 2,
71
71
  "learning_rate": 0.0058424016622,
@@ -76,12 +76,12 @@ hyperparameter_portfolio_zeroshot_2025_small = {
76
76
  "subsample": 0.809527841437,
77
77
  },
78
78
  {
79
- "ag_args": {'name_suffix': '_r10', 'priority': -12},
80
- "boosting_type": 'Plain',
81
- "bootstrap_type": 'Bernoulli',
79
+ "ag_args": {"name_suffix": "_r10", "priority": -12},
80
+ "boosting_type": "Plain",
81
+ "bootstrap_type": "Bernoulli",
82
82
  "colsample_bylevel": 0.8994502668431,
83
83
  "depth": 6,
84
- "grow_policy": 'Depthwise',
84
+ "grow_policy": "Depthwise",
85
85
  "l2_leaf_reg": 1.8187025215896,
86
86
  "leaf_estimation_iterations": 7,
87
87
  "learning_rate": 0.005177304142,
@@ -92,12 +92,12 @@ hyperparameter_portfolio_zeroshot_2025_small = {
92
92
  "subsample": 0.8705228845742,
93
93
  },
94
94
  {
95
- "ag_args": {'name_suffix': '_r24', 'priority': -15},
96
- "boosting_type": 'Plain',
97
- "bootstrap_type": 'Bernoulli',
95
+ "ag_args": {"name_suffix": "_r24", "priority": -15},
96
+ "boosting_type": "Plain",
97
+ "bootstrap_type": "Bernoulli",
98
98
  "colsample_bylevel": 0.8597809376276,
99
99
  "depth": 8,
100
- "grow_policy": 'Depthwise',
100
+ "grow_policy": "Depthwise",
101
101
  "l2_leaf_reg": 0.3628261923976,
102
102
  "leaf_estimation_iterations": 5,
103
103
  "learning_rate": 0.016851077771,
@@ -108,12 +108,12 @@ hyperparameter_portfolio_zeroshot_2025_small = {
108
108
  "subsample": 0.8120271122061,
109
109
  },
110
110
  {
111
- "ag_args": {'name_suffix': '_r91', 'priority': -17},
112
- "boosting_type": 'Plain',
113
- "bootstrap_type": 'Bernoulli',
111
+ "ag_args": {"name_suffix": "_r91", "priority": -17},
112
+ "boosting_type": "Plain",
113
+ "bootstrap_type": "Bernoulli",
114
114
  "colsample_bylevel": 0.8959275863514,
115
115
  "depth": 4,
116
- "grow_policy": 'SymmetricTree',
116
+ "grow_policy": "SymmetricTree",
117
117
  "l2_leaf_reg": 0.0026915894253,
118
118
  "leaf_estimation_iterations": 12,
119
119
  "learning_rate": 0.0475233791203,
@@ -126,10 +126,10 @@ hyperparameter_portfolio_zeroshot_2025_small = {
126
126
  ],
127
127
  "TABM": [
128
128
  {
129
- "ag_args": {'name_suffix': '_r184', 'priority': -6},
129
+ "ag_args": {"name_suffix": "_r184", "priority": -6},
130
130
  "amp": False,
131
- "arch_type": 'tabm-mini',
132
- "batch_size": 'auto',
131
+ "arch_type": "tabm-mini",
132
+ "batch_size": "auto",
133
133
  "d_block": 864,
134
134
  "d_embedding": 24,
135
135
  "dropout": 0.0,
@@ -137,17 +137,17 @@ hyperparameter_portfolio_zeroshot_2025_small = {
137
137
  "lr": 0.0019256819924656217,
138
138
  "n_blocks": 3,
139
139
  "num_emb_n_bins": 3,
140
- "num_emb_type": 'pwl',
140
+ "num_emb_type": "pwl",
141
141
  "patience": 16,
142
142
  "share_training_batches": False,
143
143
  "tabm_k": 32,
144
144
  "weight_decay": 0.0,
145
145
  },
146
146
  {
147
- "ag_args": {'name_suffix': '_r69', 'priority': -7},
147
+ "ag_args": {"name_suffix": "_r69", "priority": -7},
148
148
  "amp": False,
149
- "arch_type": 'tabm-mini',
150
- "batch_size": 'auto',
149
+ "arch_type": "tabm-mini",
150
+ "batch_size": "auto",
151
151
  "d_block": 848,
152
152
  "d_embedding": 28,
153
153
  "dropout": 0.40215621636031007,
@@ -155,17 +155,17 @@ hyperparameter_portfolio_zeroshot_2025_small = {
155
155
  "lr": 0.0010413640454559532,
156
156
  "n_blocks": 3,
157
157
  "num_emb_n_bins": 18,
158
- "num_emb_type": 'pwl',
158
+ "num_emb_type": "pwl",
159
159
  "patience": 16,
160
160
  "share_training_batches": False,
161
161
  "tabm_k": 32,
162
162
  "weight_decay": 0.0,
163
163
  },
164
164
  {
165
- "ag_args": {'name_suffix': '_r52', 'priority': -11},
165
+ "ag_args": {"name_suffix": "_r52", "priority": -11},
166
166
  "amp": False,
167
- "arch_type": 'tabm-mini',
168
- "batch_size": 'auto',
167
+ "arch_type": "tabm-mini",
168
+ "batch_size": "auto",
169
169
  "d_block": 1024,
170
170
  "d_embedding": 32,
171
171
  "dropout": 0.0,
@@ -173,20 +173,20 @@ hyperparameter_portfolio_zeroshot_2025_small = {
173
173
  "lr": 0.0006297851297842611,
174
174
  "n_blocks": 4,
175
175
  "num_emb_n_bins": 22,
176
- "num_emb_type": 'pwl',
176
+ "num_emb_type": "pwl",
177
177
  "patience": 16,
178
178
  "share_training_batches": False,
179
179
  "tabm_k": 32,
180
180
  "weight_decay": 0.06900108498839816,
181
181
  },
182
182
  {
183
- "ag_args": {'priority': -13},
183
+ "ag_args": {"priority": -13},
184
184
  },
185
185
  {
186
- "ag_args": {'name_suffix': '_r191', 'priority': -14},
186
+ "ag_args": {"name_suffix": "_r191", "priority": -14},
187
187
  "amp": False,
188
- "arch_type": 'tabm-mini',
189
- "batch_size": 'auto',
188
+ "arch_type": "tabm-mini",
189
+ "batch_size": "auto",
190
190
  "d_block": 864,
191
191
  "d_embedding": 8,
192
192
  "dropout": 0.45321529282058803,
@@ -194,17 +194,17 @@ hyperparameter_portfolio_zeroshot_2025_small = {
194
194
  "lr": 0.0003781238075322413,
195
195
  "n_blocks": 4,
196
196
  "num_emb_n_bins": 27,
197
- "num_emb_type": 'pwl',
197
+ "num_emb_type": "pwl",
198
198
  "patience": 16,
199
199
  "share_training_batches": False,
200
200
  "tabm_k": 32,
201
201
  "weight_decay": 0.01766851962579851,
202
202
  },
203
203
  {
204
- "ag_args": {'name_suffix': '_r49', 'priority': -20},
204
+ "ag_args": {"name_suffix": "_r49", "priority": -20},
205
205
  "amp": False,
206
- "arch_type": 'tabm-mini',
207
- "batch_size": 'auto',
206
+ "arch_type": "tabm-mini",
207
+ "batch_size": "auto",
208
208
  "d_block": 640,
209
209
  "d_embedding": 28,
210
210
  "dropout": 0.15296207419190627,
@@ -212,7 +212,7 @@ hyperparameter_portfolio_zeroshot_2025_small = {
212
212
  "lr": 0.002277678490593717,
213
213
  "n_blocks": 3,
214
214
  "num_emb_n_bins": 48,
215
- "num_emb_type": 'pwl',
215
+ "num_emb_type": "pwl",
216
216
  "patience": 16,
217
217
  "share_training_batches": False,
218
218
  "tabm_k": 32,
@@ -221,16 +221,16 @@ hyperparameter_portfolio_zeroshot_2025_small = {
221
221
  ],
222
222
  "TABICL": [
223
223
  {
224
- "ag_args": {'priority': -8},
224
+ "ag_args": {"priority": -8},
225
225
  },
226
226
  ],
227
227
  "XGB": [
228
228
  {
229
- "ag_args": {'name_suffix': '_r171', 'priority': -9},
229
+ "ag_args": {"name_suffix": "_r171", "priority": -9},
230
230
  "colsample_bylevel": 0.9213705632288,
231
231
  "colsample_bynode": 0.6443385965381,
232
232
  "enable_categorical": True,
233
- "grow_policy": 'lossguide',
233
+ "grow_policy": "lossguide",
234
234
  "learning_rate": 0.0068171645251,
235
235
  "max_cat_to_onehot": 8,
236
236
  "max_depth": 6,
@@ -241,11 +241,11 @@ hyperparameter_portfolio_zeroshot_2025_small = {
241
241
  "subsample": 0.9656290596647,
242
242
  },
243
243
  {
244
- "ag_args": {'name_suffix': '_r40', 'priority': -18},
244
+ "ag_args": {"name_suffix": "_r40", "priority": -18},
245
245
  "colsample_bylevel": 0.6377491713202,
246
246
  "colsample_bynode": 0.9237625621103,
247
247
  "enable_categorical": True,
248
- "grow_policy": 'lossguide',
248
+ "grow_policy": "lossguide",
249
249
  "learning_rate": 0.0112462621131,
250
250
  "max_cat_to_onehot": 33,
251
251
  "max_depth": 10,
@@ -262,7 +262,7 @@ hyperparameter_portfolio_zeroshot_2025_small = {
262
262
  "fine_tune": True,
263
263
  "fine_tune_steps": 50,
264
264
  "ag.num_gpus": 1,
265
- "ag_args": {'priority': -21},
265
+ "ag_args": {"priority": -21},
266
266
  },
267
267
  ],
268
268
  }