|
| 1 | +{ |
| 2 | + "INCLUDE": ["../common/sklearn.json"], |
| 3 | + "PARAMETERS_SETS": { |
| 4 | + "throughput settings": { |
| 5 | + "bench": { |
| 6 | + "num_instances": 7, |
| 7 | + "cores_per_instance": 8, |
| 8 | + "measurement_duration": 30 |
| 9 | + } |
| 10 | + }, |
| 11 | + "linear regression": { |
| 12 | + "algorithm": { |
| 13 | + "estimator": "LinearRegression", |
| 14 | + "estimator_params": { |
| 15 | + "fit_intercept": true, |
| 16 | + "copy_X": true |
| 17 | + } |
| 18 | + } |
| 19 | + }, |
| 20 | + "random forest classifier": { |
| 21 | + "algorithm": { |
| 22 | + "estimator": "RandomForestClassifier", |
| 23 | + "estimator_params": { |
| 24 | + "n_estimators": 500, |
| 25 | + "max_depth": 12, |
| 26 | + "max_samples": 0.8, |
| 27 | + "min_samples_split": 5, |
| 28 | + "min_samples_leaf": 2, |
| 29 | + "bootstrap": true, |
| 30 | + "criterion": "gini", |
| 31 | + "max_features": "sqrt", |
| 32 | + "random_state": 42 |
| 33 | + } |
| 34 | + } |
| 35 | + }, |
| 36 | + "knn regressor kdtree": { |
| 37 | + "algorithm": { |
| 38 | + "estimator": "KNeighborsRegressor", |
| 39 | + "estimator_params": { |
| 40 | + "n_neighbors": 100, |
| 41 | + "weights": "uniform", |
| 42 | + "algorithm": "kd_tree", |
| 43 | + "metric": "minkowski", |
| 44 | + "p": 2 |
| 45 | + } |
| 46 | + }, |
| 47 | + "data": { |
| 48 | + "preprocessing_kwargs": { "normalize": "standard" } |
| 49 | + } |
| 50 | + }, |
| 51 | + "kmeans": { |
| 52 | + "algorithm": { |
| 53 | + "estimator": "KMeans", |
| 54 | + "estimator_params": { |
| 55 | + "n_clusters": 10, |
| 56 | + "n_init": 1, |
| 57 | + "max_iter": 30, |
| 58 | + "tol": 1e-3, |
| 59 | + "random_state": 42, |
| 60 | + "init": "k-means++", |
| 61 | + "algorithm": "lloyd" |
| 62 | + }, |
| 63 | + "estimator_methods": { |
| 64 | + "inference": "predict" |
| 65 | + } |
| 66 | + }, |
| 67 | + "data": { |
| 68 | + "preprocessing_kwargs": { "normalize": "standard" } |
| 69 | + } |
| 70 | + }, |
| 71 | + "linear regression datasets": { |
| 72 | + "data": [ |
| 73 | + { |
| 74 | + "dataset": "hepmass", |
| 75 | + "split_kwargs": { "train_size": 2000000, "test_size": null } |
| 76 | + }, |
| 77 | + { |
| 78 | + "dataset": "year_prediction_msd", |
| 79 | + "split_kwargs": { "train_size": 0.5, "test_size": 0.5 } |
| 80 | + } |
| 81 | + ] |
| 82 | + }, |
| 83 | + "random forest classifier datasets": { |
| 84 | + "data": [ |
| 85 | + { |
| 86 | + "dataset": "codrnanorm", |
| 87 | + "split_kwargs": { "train_size": 0.25, "test_size": null } |
| 88 | + }, |
| 89 | + { |
| 90 | + "dataset": "connect", |
| 91 | + "split_kwargs": { "train_size": 0.5, "test_size": null } |
| 92 | + } |
| 93 | + ] |
| 94 | + }, |
| 95 | + "knn regressor datasets": { |
| 96 | + "data": [ |
| 97 | + { |
| 98 | + "dataset": "skin_segmentation", |
| 99 | + "split_kwargs": { "train_size": 0.25, "test_size": 0.75 } |
| 100 | + }, |
| 101 | + { |
| 102 | + "dataset": "medical_charges_nominal", |
| 103 | + "split_kwargs": { "train_size": 0.75, "test_size": 0.25 } |
| 104 | + } |
| 105 | + ] |
| 106 | + }, |
| 107 | + "kmeans datasets": [ |
| 108 | + { |
| 109 | + "data": { |
| 110 | + "dataset": "hepmass", |
| 111 | + "split_kwargs": { |
| 112 | + "train_size": 2000000, |
| 113 | + "test_size": null, |
| 114 | + "shuffle": true, |
| 115 | + "random_state": 42 |
| 116 | + } |
| 117 | + }, |
| 118 | + "algorithm": { |
| 119 | + "estimator_params": { "n_clusters": 50 } |
| 120 | + } |
| 121 | + }, |
| 122 | + { |
| 123 | + "data": { |
| 124 | + "dataset": "mnist", |
| 125 | + "split_kwargs": { "ignore": true } |
| 126 | + } |
| 127 | + } |
| 128 | + ] |
| 129 | + }, |
| 130 | + "TEMPLATES": { |
| 131 | + "linear_regression_throughput": { |
| 132 | + "SETS": [ |
| 133 | + "sklearn-ex[cpu] implementations", |
| 134 | + "throughput settings", |
| 135 | + "linear regression", |
| 136 | + "linear regression datasets" |
| 137 | + ] |
| 138 | + }, |
| 139 | + "random_forest_classifier_throughput": { |
| 140 | + "SETS": [ |
| 141 | + "sklearn-ex[cpu] implementations", |
| 142 | + "throughput settings", |
| 143 | + "random forest classifier", |
| 144 | + "random forest classifier datasets" |
| 145 | + ] |
| 146 | + }, |
| 147 | + "knn_regressor_throughput": { |
| 148 | + "SETS": [ |
| 149 | + "sklearn-ex[cpu] implementations", |
| 150 | + "throughput settings", |
| 151 | + "knn regressor kdtree", |
| 152 | + "knn regressor datasets" |
| 153 | + ] |
| 154 | + }, |
| 155 | + "kmeans_throughput": { |
| 156 | + "SETS": [ |
| 157 | + "sklearn-ex[cpu] implementations", |
| 158 | + "throughput settings", |
| 159 | + "kmeans", |
| 160 | + "kmeans datasets" |
| 161 | + ] |
| 162 | + } |
| 163 | + } |
| 164 | +} |
0 commit comments