This example runs Skip Feature Engineering to perform categorical encoding manually before fit. Run AutoML to get the best performing model with the following specifications:
- Remove duplicates and drop redundant columns manually if required.
- Drop or impute missing values.
- Encode categorical columns
- Load the online fraud data.
>>> load_example_data('teradataml', ['payment_fraud_dataset'])>>> fraud = DataFrame.from_table("payment_fraud_dataset") >>> df = DataFrame('payment_fraud_dataset') >>> fraud_sample = df.sample(frac = [0.8, 0.2]) >>> fraud_train = fraud_sample[fraud_sample['sampleid'] == 1].drop('sampleid', axis=1) >>> fraud_test = fraud_sample[fraud_sample['sampleid'] == 2].drop('sampleid', axis=1) - Drop the missing values.
>>> fraud_train_up = fraud_train.dropna() >>> fraud_test_up = fraud_test.dropna()
- Encode categorical columns.
>>> fraud_train_up=fraud_train_up.drop(columns=['payment_type','nameOrig','nameDest'], axis=1) >>> fraud_train_up
amount oldbalanceOrg newbalanceOrig oldbalanceDest newbalanceDest isFraud step 38 4221.67 34153.28 29931.61 0.00 0.00 0 17 8500.03 110230.59 101730.56 0.00 0.00 0 17 35859.13 0.00 0.00 322079.03 357938.15 0 19 287005.80 5063.00 0.00 102335.41 389341.21 0 19 228341.16 0.00 0.00 392457.68 620798.84 0 40 75672.79 0.00 0.00 478121.23 553794.01 0 40 120549.52 0.00 0.00 2381888.19 2502437.71 0 40 336996.80 0.00 0.00 419362.17 756358.97 0 19 199049.77 0.00 0.00 263051.65 467932.04 0 17 251240.86 52381.00 0.00 4248406.53 4499647.39 0
>>> fraud_test_up=fraud_test_up.drop(columns=['payment_type','nameOrig','nameDest'], axis=1) >>> fraud_test_up
amount oldbalanceOrg newbalanceOrig oldbalanceDest newbalanceDest isFraud step 19 219928.90 0.00 0.0 437186.91 657115.81 0 38 756542.05 113879.00 0.0 110973.66 867515.71 0 38 360688.30 360688.30 0.0 0.00 360688.30 1 40 105987.63 0.00 0.0 1234821.83 1340809.46 0 40 9008.48 0.00 0.0 0.00 0.00 0 17 304755.42 0.00 0.0 2079910.10 2384665.52 0 17 509836.85 0.00 0.0 2933460.49 3443297.35 0 17 198711.10 225234.00 26522.9 1417025.29 1615736.39 0 40 482260.45 0.00 0.0 6848237.76 7330498.21 0 61 475368.94 475368.94 0.0 1348026.73 1823395.67 1
- Create an AutoML instance.
>>> aml = AutoFraud(max_runtime_secs=36, >>> stopping_metric="Weighted-F1", >>> stopping_tolerance=0.30, >>> verbose=2, >>> skip_phases='Feature_Engineering')
- Fit the data.
>>> aml.fit(fraud_train_up, fraud_train_up.isFraud)
1. Feature Exploration -> 2. Data Preparation -> 3. Model Training & Evaluation% - 0/10 2026-04-05 00:31:06,702 | INFO | Feature Exploration started 2026-04-05 00:31:06,702 | INFO | Data Overview: 2026-04-05 00:31:06,807 | INFO | Total Rows in the data: 7960 2026-04-05 00:31:06,808 | INFO | Total Columns in the data: 7 2026-04-05 00:31:07,869 | INFO | Column Summary: ColumnName Datatype NonNullCount NullCount BlankCount ZeroCount PositiveCount NegativeCount NullPercentage NonNullPercentage 0 newbalanceOrig FLOAT 7960 0 None 3977 3983 0 0.0 100.0 1 step BIGINT 7960 0 None 0 7960 0 0.0 100.0 2 isFraud BIGINT 7960 0 None 7804 156 0 0.0 100.0 3 oldbalanceDest FLOAT 7960 0 None 3957 4003 0 0.0 100.0 4 newbalanceDest FLOAT 7960 0 None 3857 4103 0 0.0 100.0 5 oldbalanceOrg FLOAT 7960 0 None 2246 5714 0 0.0 100.0 6 amount FLOAT 7960 0 None 0 7960 0 0.0 100.0 2026-04-05 00:31:09,375 | INFO | Statistics of Data: ATTRIBUTE StatName StatValue 0 newbalanceDest MAXIMUM 34600000.00 1 isFraud MINIMUM 0.00 2 isFraud MAXIMUM 1.00 3 newbalanceOrig COUNT 7960.00 4 newbalanceOrig MAXIMUM 13000000.00 5 amount COUNT 7960.00 6 amount MINIMUM 0.65 7 amount MAXIMUM 10000000.00 8 newbalanceOrig MINIMUM 0.00 9 isFraud COUNT 7960.00 2026-04-05 00:31:12,604 | INFO | Columns with outlier percentage :- ColumnName OutlierPercentage 0 oldbalanceOrg 0.992462 1 newbalanceDest 0.992462 2 newbalanceOrig 0.992462 3 amount 1.984925 4 oldbalanceDest 0.954774 2026-04-05 00:31:12,891 | INFO | Skipping Feature_Engineering phase as requested. 1. Feature Exploration -> 2. Data Preparation -> 3. Model Training & Evaluation 2026-04-05 00:31:13,278 | INFO | Data preparation started ... 2026-04-05 00:31:13,278 | INFO | AutoFraud Outlier preprocessing using Percentile... 2026-04-05 00:31:16,156 | INFO | Columns with outlier percentage :- ColumnName OutlierPercentage 0 oldbalanceOrg 0.992462 1 newbalanceDest 0.992462 2 newbalanceOrig 0.992462 3 amount 1.984925 4 oldbalanceDest 0.954774 2026-04-05 00:31:16,929 | INFO | Replacing outliers with median: ['amount', 'oldbalanceDest', 'oldbalanceOrg', 'newbalanceOrig', 'newbalanceDest'] 2026-04-05 00:31:18,778 | INFO | Sample of dataset after replacing outliers with MEDIAN: amount oldbalanceOrg newbalanceOrig oldbalanceDest newbalanceDest isFraud automl_id step 40 31148.345 0.00 0.00 1856992.92 2521238.14 0 170 40 262201.570 0.00 0.00 3134400.04 3396601.61 0 294 40 66864.600 0.00 0.00 507073.29 573937.89 0 498 40 31148.345 0.00 0.00 0.00 0.00 0 246 40 322054.450 10617.00 0.00 308801.79 630856.24 0 534 40 84002.970 0.00 0.00 838741.88 922744.85 0 194 40 401413.270 0.00 0.00 948547.60 1349960.87 0 470 40 70186.080 10771.00 0.00 821430.71 891616.79 0 114 40 15943.720 18470.14 2526.42 0.00 0.00 0 402 40 91654.190 0.00 0.00 694270.40 785924.59 0 118 7960 rows X 8 columns 2026-04-05 00:31:18,895 | INFO | Time Taken by Outlier processing: 5.62 sec 2026-04-05 00:31:18,896 | INFO | Checking imbalance data ... 2026-04-05 00:31:18,959 | INFO | Imbalance Found. 2026-04-05 00:31:18,960 | INFO | Handling data imbalance using SMOTE ... 2026-04-05 00:31:22,213 | INFO | Completed data imbalance handling. 2026-04-05 00:31:22,580 | INFO | Feature selection using rfe ... File lasso_rfe.py replaced in Vantage________________________________________| 30% - 3/10 2026-04-05 00:31:45,270 | INFO | feature selected by rfe: ['step', 'amount', 'oldbalanceOrg', 'newbalanceOrig', 'oldbalanceDest', 'newbalanceDest'] 2026-04-05 00:31:45,355 | INFO | Total time taken by feature selection: 22.78 sec 2026-04-05 00:31:45,355 | INFO | Scaling Features of rfe data ... 2026-04-05 00:31:46,471 | INFO | columns that will be scaled: ['r_amount', 'r_newbalanceDest', 'r_newbalanceOrig', 'r_oldbalanceDest', 'r_oldbalanceOrg', 'r_step'] 2026-04-05 00:31:48,824 | INFO | Dataset sample after scaling: automl_id isFraud r_amount r_newbalanceDest r_newbalanceOrig r_oldbalanceDest r_oldbalanceOrg r_step 0 2345 0 0.019673 0.062156 0.034073 0.067889 0.036727 0.510638 1 3487 0 0.045726 0.000000 0.008120 0.000000 0.013582 0.053191 2 3956 0 0.293436 0.016737 0.000000 0.062102 0.000000 0.021277 3 3018 0 0.054619 0.302106 0.000000 0.053402 0.024949 0.382979 4 938 0 0.010425 0.000000 0.000845 0.000000 0.002259 1.000000 5 469 0 0.039167 0.000000 0.007689 0.000000 0.012404 0.170213 6 5832 0 0.042382 0.000000 0.000000 0.000000 0.000000 0.446809 7 11868 0 0.157465 0.180039 0.000000 0.186548 0.000000 0.010638 8 12072 0 0.017522 0.000000 0.008420 0.000000 0.010683 0.010638 9 6036 0 0.096431 0.349997 0.000000 0.346970 0.000000 0.446809 8740 rows X 8 columns 2026-04-05 00:31:49,467 | INFO | Total time taken by feature scaling: 4.11 sec 2026-04-05 00:31:49,468 | INFO | Scaling Features of pca data ... 2026-04-05 00:31:50,065 | INFO | columns that will be scaled: ['step', 'amount', 'oldbalanceOrg', 'newbalanceOrig', 'oldbalanceDest', 'newbalanceDest'] 2026-04-05 00:31:51,117 | INFO | Dataset sample after scaling: automl_id isFraud step amount oldbalanceOrg newbalanceOrig oldbalanceDest newbalanceDest 0 265 0 0.170213 0.421468 0.005433 0.053186 0.000202 0.941635 1 307 0 0.191489 0.414908 0.030142 0.000000 0.499831 0.482250 2 263 0 0.191489 0.013948 0.145398 0.142771 0.000000 0.000000 3 390 0 0.414894 0.038024 0.000000 0.000000 0.000000 0.000000 4 50 0 0.414894 0.054606 0.009652 0.015984 0.000000 0.000000 5 12404 1 0.638298 0.066215 0.395540 0.000000 0.000000 0.000000 6 12400 1 0.638298 0.101405 0.014314 0.000000 0.000000 0.006979 7 12420 1 0.638298 0.446383 0.046783 0.000000 0.009869 0.173823 8 530 0 0.414894 0.054619 0.001259 0.001231 0.403165 0.357233 9 119 0 0.191489 0.082450 0.000000 0.000000 0.685716 0.615376 8740 rows X 8 columns 2026-04-05 00:31:51,697 | INFO | Total time taken by feature scaling: 2.23 sec 2026-04-05 00:31:51,910 | INFO | Dimension Reduction using pca ... File pca_fs.py replaced in Vantage███████____________________________________| 40% - 4/10 2026-04-05 00:32:01,633 | INFO | PCA columns: ['col_0', 'col_1', 'col_2', 'col_3'] 2026-04-05 00:32:01,634 | INFO | Total time taken by PCA: 9.72 sec 1. Feature Exploration -> 2. Data Preparation -> 3. Model Training & Evaluation 2026-04-05 00:32:01,727 | INFO | Model Training started ... 2026-04-05 00:32:01,770 | INFO | Hyperparameters used for model training: 2026-04-05 00:32:01,770 | INFO | Model: glm 2026-04-05 00:32:01,770 | INFO | Hyperparameters: {'response_column': 'isFraud', 'name': 'glm', 'family': 'BINOMIAL', 'lambda1': (0.001, 0.02, 0.1), 'alpha': (0.15, 0.85), 'learning_rate': 'OPTIMAL', 'initial_eta': (0.05, 0.1), 'momentum': (0.65, 0.8, 0.95), 'iter_num_no_change': (5, 10, 50), 'iter_max': (300, 200, 400), 'batch_size': (10, 50, 60, 80)} 2026-04-05 00:32:01,771 | INFO | Total number of models for glm: 1296 -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 2026-04-05 00:32:01,771 | INFO | Model: svm 2026-04-05 00:32:01,772 | INFO | Hyperparameters: {'response_column': 'isFraud', 'name': 'svm', 'model_type': 'Classification', 'lambda1': (0.001, 0.02, 0.1), 'alpha': (0.15, 0.85), 'tolerance': (0.001, 0.01), 'learning_rate': 'OPTIMAL', 'initial_eta': (0.05, 0.1), 'momentum': (0.65, 0.8, 0.95), 'nesterov': True, 'intercept': True, 'iter_num_no_change': (5, 10, 50), 'local_sgd_iterations ': (10, 20), 'iter_max': (300, 200, 400), 'batch_size': (10, 50, 60, 80)} 2026-04-05 00:32:01,774 | INFO | Total number of models for svm: 5184 -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 2026-04-05 00:32:01,774 | INFO | Model: knn 2026-04-05 00:32:01,775 | INFO | Hyperparameters: {'response_column': 'isFraud', 'name': 'knn', 'model_type': 'Classification', 'k': (3, 5, 6, 8, 10, 12), 'id_column': 'automl_id', 'voting_weight': 1.0} 2026-04-05 00:32:01,775 | INFO | Total number of models for knn: 6 -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 2026-04-05 00:32:01,775 | INFO | Model: decision_forest 2026-04-05 00:32:01,775 | INFO | Hyperparameters: {'response_column': 'isFraud', 'name': 'decision_forest', 'tree_type': 'Classification', 'min_impurity': (0.0, 0.1, 0.2), 'max_depth': (5, 6, 8, 10), 'min_node_size': (1, 2, 3), 'num_trees': (-1,), 'seed': 42} 2026-04-05 00:32:01,775 | INFO | Total number of models for decision_forest: 36 -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 2026-04-05 00:32:01,775 | INFO | Model: xgboost 2026-04-05 00:32:01,775 | INFO | Hyperparameters: {'response_column': 'isFraud', 'name': 'xgboost', 'model_type': 'Classification', 'column_sampling': (1, 0.6), 'min_impurity': (0.0, 0.1, 0.2), 'lambda1': (1.0, 0.01, 0.1), 'shrinkage_factor': (0.5, 0.1, 0.3), 'max_depth': (5, 6, 8, 10), 'min_node_size': (1, 2, 3), 'iter_num': (10, 20, 30), 'num_boosted_trees': (-1, 5, 10), 'seed': 42} 2026-04-05 00:32:01,777 | INFO | Total number of models for xgboost: 5832 -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 2026-04-05 00:32:01,777 | INFO | Performing hyperparameter tuning ... 2026-04-05 00:32:03,175 | INFO | Model training for glm 2026-04-05 00:32:15,629 | INFO | ---------------------------------------------------------------------------------------------------- 2026-04-05 00:32:15,630 | INFO | Model training for svm 2026-04-05 00:32:24,602 | INFO | ---------------------------------------------------------------------------------------------------- 2026-04-05 00:32:24,602 | INFO | Model training for knn 2026-04-05 00:33:11,450 | INFO | ---------------------------------------------------------------------------------------------------- 2026-04-05 00:33:11,451 | INFO | Model training for decision_forest 2026-04-05 00:33:26,774 | INFO | ---------------------------------------------------------------------------------------------------- 2026-04-05 00:33:26,775 | INFO | Model training for xgboost 2026-04-05 00:33:37,874 | INFO | ---------------------------------------------------------------------------------------------------- 2026-04-05 00:33:37,878 | INFO | Leaderboard RANK MODEL_ID FEATURE_SELECTION ACCURACY ... MACRO-F1 WEIGHTED-PRECISION WEIGHTED-RECALL WEIGHTED-F1 0 1 XGBOOST_0 rfe 0.971396 ... 0.921026 0.970628 0.971396 0.970530 1 2 XGBOOST_2 rfe 0.970252 ... 0.916602 0.969499 0.970252 0.969104 2 3 DECISIONFOREST_2 rfe 0.967963 ... 0.906308 0.967872 0.967963 0.965963 3 4 DECISIONFOREST_0 rfe 0.967963 ... 0.905796 0.968067 0.967963 0.965862 4 5 KNN_4 rfe 0.958238 ... 0.873376 0.957782 0.958238 0.954739 5 6 KNN_0 rfe 0.955950 ... 0.869397 0.954340 0.955950 0.952848 6 7 KNN_3 pca 0.939931 ... 0.823848 0.935673 0.939931 0.936086 7 8 KNN_7 pca 0.938215 ... 0.805451 0.934095 0.938215 0.931595 8 9 XGBOOST_3 pca 0.924485 ... 0.743763 0.917521 0.924485 0.912667 9 10 XGBOOST_1 pca 0.920481 ... 0.760256 0.912251 0.920481 0.914089 10 11 DECISIONFOREST_3 pca 0.917048 ... 0.715745 0.906577 0.917048 0.903500 11 12 DECISIONFOREST_1 pca 0.917048 ... 0.713855 0.906716 0.917048 0.903116 12 13 GLM_0 rfe 0.892449 ... 0.471584 0.796464 0.892449 0.841729 13 14 GLM_3 pca 0.892449 ... 0.471584 0.796464 0.892449 0.841729 14 15 GLM_2 rfe 0.892449 ... 0.471584 0.796464 0.892449 0.841729 15 16 GLM_1 pca 0.892449 ... 0.471584 0.796464 0.892449 0.841729 16 17 SVM_0 rfe 0.892449 ... 0.471584 0.796464 0.892449 0.841729 17 18 SVM_2 rfe 0.892449 ... 0.471584 0.796464 0.892449 0.841729 18 19 SVM_1 pca 0.892449 ... 0.471584 0.796464 0.892449 0.841729 19 20 SVM_3 pca 0.892449 ... 0.471584 0.796464 0.892449 0.841729 [20 rows x 13 columns] 20 rows X 13 columns 1. Feature Exploration -> 2. Data Preparation -> 3. Model Training & Evaluation Completed: ｜⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿⫿｜ 100% - 10/10 - Display model leaderboard.
>>> aml.leaderboard()
RANK MODEL_ID FEATURE_SELECTION ACCURACY ... MACRO-F1 WEIGHTED-PRECISION WEIGHTED-RECALL WEIGHTED-F1 0 1 XGBOOST_0 rfe 0.971396 ... 0.921026 0.970628 0.971396 0.970530 1 2 XGBOOST_2 rfe 0.970252 ... 0.916602 0.969499 0.970252 0.969104 2 3 DECISIONFOREST_2 rfe 0.967963 ... 0.906308 0.967872 0.967963 0.965963 3 4 DECISIONFOREST_0 rfe 0.967963 ... 0.905796 0.968067 0.967963 0.965862 4 5 KNN_4 rfe 0.958238 ... 0.873376 0.957782 0.958238 0.954739 5 6 KNN_0 rfe 0.955950 ... 0.869397 0.954340 0.955950 0.952848 6 7 KNN_3 pca 0.939931 ... 0.823848 0.935673 0.939931 0.936086 7 8 KNN_7 pca 0.938215 ... 0.805451 0.934095 0.938215 0.931595 8 9 XGBOOST_3 pca 0.924485 ... 0.743763 0.917521 0.924485 0.912667 9 10 XGBOOST_1 pca 0.920481 ... 0.760256 0.912251 0.920481 0.914089 10 11 DECISIONFOREST_3 pca 0.917048 ... 0.715745 0.906577 0.917048 0.903500 11 12 DECISIONFOREST_1 pca 0.917048 ... 0.713855 0.906716 0.917048 0.903116 12 13 GLM_0 rfe 0.892449 ... 0.471584 0.796464 0.892449 0.841729 13 14 GLM_3 pca 0.892449 ... 0.471584 0.796464 0.892449 0.841729 14 15 GLM_2 rfe 0.892449 ... 0.471584 0.796464 0.892449 0.841729 15 16 GLM_1 pca 0.892449 ... 0.471584 0.796464 0.892449 0.841729 16 17 SVM_0 rfe 0.892449 ... 0.471584 0.796464 0.892449 0.841729 17 18 SVM_2 rfe 0.892449 ... 0.471584 0.796464 0.892449 0.841729 18 19 SVM_1 pca 0.892449 ... 0.471584 0.796464 0.892449 0.841729 19 20 SVM_3 pca 0.892449 ... 0.471584 0.796464 0.892449 0.841729 [20 rows x 13 columns]
- Display best performing model.
>>> aml.leader()
RANK MODEL_ID FEATURE_SELECTION ACCURACY MICRO-PRECISION ... MACRO-RECALL MACRO-F1 WEIGHTED-PRECISION WEIGHTED-RECALL WEIGHTED-F1 0 1 XGBOOST_0 rfe 0.971396 0.971396 ... 0.897429 0.921026 0.970628 0.971396 0.97053 [1 rows x 13 columns]
- Display model hyperparameters for rank 1.
>>> aml.model_hyperparameters(rank=1)
{'response_column': 'isFraud', 'model_type': 'Classification', 'column_sampling': 1, 'min_impurity': 0.0, 'lambda1': 1.0, 'shrinkage_factor': 0.5, 'max_depth': 5, 'min_node_size': 1, 'iter_num': 10, 'num_boosted_trees': -1, 'seed': 42, 'output_prob': True, 'output_responses': ['1', '0']} - Generate prediction on test dataset using best performing model.
>>> prediction = aml.predict(fraud_test_up)
2026-04-05 00:41:48,049 | INFO | Data Transformation started ... 2026-04-05 00:41:48,050 | INFO | Performing transformation carried out in feature engineering phase ... 2026-04-05 00:41:48,467 | INFO | Updated dataset after performing target column transformation : step amount oldbalanceOrg newbalanceOrig oldbalanceDest newbalanceDest isFraud automl_id 0 38 360688.30 360688.3 0.0 0.00 360688.30 1 12 1 17 509836.85 0.0 0.0 2933460.49 3443297.35 0 9 2 17 198711.10 225234.0 26522.9 1417025.29 1615736.39 0 13 3 40 105987.63 0.0 0.0 1234821.83 1340809.46 0 6 4 40 9008.48 0.0 0.0 0.00 0.00 0 14 5 19 26929.48 0.0 0.0 971058.99 997988.47 0 7 6 19 133681.99 14837.0 0.0 0.00 133681.99 0 11 7 19 219928.90 0.0 0.0 437186.91 657115.81 0 15 8 40 482260.45 0.0 0.0 6848237.76 7330498.21 0 10 9 17 304755.42 0.0 0.0 2079910.10 2384665.52 0 5 1990 rows X 8 columns 2026-04-05 00:41:49,087 | INFO | Performing transformation carried out in data preparation phase ... 2026-04-05 00:41:49,835 | INFO | Updated dataset after performing scaling for PCA feature selection : automl_id isFraud step amount oldbalanceOrg newbalanceOrig oldbalanceDest newbalanceDest 0 326 0 0.063830 0.170151 1.773182 1.782513 0.057005 0.152241 1 265 0 0.351064 0.437215 0.000000 0.000000 0.555509 0.484515 2 530 0 0.223404 -0.001283 0.000008 0.000000 0.000000 0.000000 3 1407 1 0.904255 0.129864 0.014964 0.000000 0.015132 0.025607 4 1672 0 0.000000 0.008471 0.006725 0.005506 0.039797 0.002858 5 938 0 0.361702 0.195528 0.014105 0.000000 0.011247 0.028239 6 1203 0 0.053191 0.108260 0.438281 0.448228 0.011556 0.000000 7 122 0 0.414894 0.007631 0.000000 0.000000 0.000000 0.000000 8 734 0 0.382979 0.027406 0.000000 0.000000 0.000000 0.000000 9 1876 0 0.000000 0.092008 0.000182 0.000000 0.011324 0.002585 990 rows X 8 columns 2026-04-05 00:41:50,198 | INFO | Updated dataset after performing PCA feature selection : automl_id col_0 col_1 col_2 col_3 isFraud 0 469 0.100540 0.172864 0.018611 -0.487603 0 1 1876 -0.251071 0.157196 -0.050036 -0.040548 0 2 1407 0.197382 -0.620255 0.039249 0.035553 1 3 938 -0.014809 -0.131157 -0.005179 -0.074603 0 4 61 -0.210724 -0.007050 -0.036051 0.017558 0 ... ... ... ... ... ... ... 1985 1224 -0.292392 0.120096 -0.044869 0.027092 0 1986 816 -0.194442 -0.077468 -0.024546 0.056364 0 1987 1081 -0.098809 0.112421 -0.021720 -0.191589 0 1988 1550 -0.291443 0.109375 -0.044474 0.034003 0 1989 2284 -0.220527 -0.021290 -0.035186 0.041881 0 [1990 rows x 6 columns] 1990 rows X 6 columns 2026-04-05 00:41:50,577 | INFO | Updated dataset after performing RFE feature selection: step amount oldbalanceOrg newbalanceOrig oldbalanceDest newbalanceDest isFraud automl_id 326 7 94712.61 8596484.44 8691197.05 300375.97 905385.44 0 1060 3 8777.15 0.00 0.00 0.00 0.00 0 2732 2 18483.96 336667.40 318183.44 0.00 0.00 0 183 36 34606.08 1664785.32 1699391.39 4779616.11 4745010.03 0 1121 48 248751.33 0.00 0.00 1771569.57 2020320.90 0 917 33 20.98 247117.26 247096.28 0.00 0.00 0 652 5 26074.81 5014.31 0.00 0.00 0.00 0 795 6 300026.34 1161124.63 1461150.97 398176.00 98149.66 0 61 17 20763.35 0.00 0.00 0.00 0.00 0 469 11 361489.71 19111.00 0.00 0.00 13277.21 0 1990 rows X 8 columns 2026-04-05 00:41:51,153 | INFO | Updated dataset after performing scaling on RFE selected features : automl_id isFraud r_amount r_newbalanceDest r_newbalanceOrig r_oldbalanceDest r_oldbalanceOrg r_step 0 122 0 0.007631 0.000000 0.000000 0.000000 0.000000 0.414894 1 734 0 0.027406 0.000000 0.000000 0.000000 0.000000 0.382979 2 1672 0 0.008471 0.002858 0.005506 0.039797 0.006725 0.000000 3 1876 0 0.092008 0.002585 0.000000 0.011324 0.000182 0.000000 4 530 0 -0.001283 0.000000 0.000000 0.000000 0.000008 0.223404 5 469 0 0.655037 0.002233 0.000000 0.000000 0.003942 0.106383 6 61 0 0.035744 0.000000 0.000000 0.000000 0.000000 0.170213 7 326 0 0.170151 0.152241 1.782513 0.057005 1.773182 0.063830 8 265 0 0.437215 0.484515 0.000000 0.555509 0.000000 0.351064 9 1407 1 0.129864 0.025607 0.000000 0.015132 0.014964 0.904255 1990 rows X 8 columns 2026-04-05 00:41:51,791 | INFO | Data Transformation completed.█████| 100% - 9/9 2026-04-05 00:41:52,342 | INFO | Following model is being picked for evaluation: 2026-04-05 00:41:52,342 | INFO | Model ID : XGBOOST_0 2026-04-05 00:41:52,343 | INFO | Feature Selection Method : rfe 2026-04-05 00:41:52,695 | INFO | Skipping SHAP analysis as enable_shap is set to False. 2026-04-05 00:41:52,748 | INFO | Prediction : 2026-04-05 00:41:52,939 | INFO | automl_id Prediction isFraud prob_0 prob_1 0 122 0 0 0.956113 0.043887 1 734 0 0 0.927625 0.072375 2 1672 0 0 0.989118 0.010882 3 469 0 0 0.775660 0.224340 4 326 0 0 0.903095 0.096905 5 1876 0 0 0.985675 0.014325 6 265 0 0 0.901958 0.098042 7 530 0 0 0.966513 0.033487 8 61 0 0 0.983711 0.016289 9 1407 1 1 0.036103 0.963897 - Generate evaluation metrics on formatted test dataset.
>>> performance_metrics = aml.evaluate(fraud_test_up)
2026-04-05 00:43:32,106 | INFO | Skipping data transformation as data is already transformed. 2026-04-05 00:43:33,258 | INFO | Following model is being picked for evaluation: 2026-04-05 00:43:33,259 | INFO | Model ID : XGBOOST_0 2026-04-05 00:43:33,259 | INFO | Feature Selection Method : rfe 2026-04-05 00:43:34,682 | INFO | Performance Metrics : 2026-04-05 00:43:34,757 | INFO | Prediction Mapping CLASS_1 CLASS_2 Precision Recall F1 Support SeqNum 1 1 CLASS_2 29 35 0.546875 0.875000 0.673077 40 0 0 CLASS_1 1921 5 0.997404 0.985128 0.991228 1950