xgboost random forest test is too slow

montanalow · montanalow · commit 00cc108e6a37 · 2023-03-15T10:56:52.000-07:00
diff --git a/pgml-extension/examples/binary_classification.sql b/pgml-extension/examples/binary_classification.sql
@@ -119,7 +119,7 @@ SELECT * FROM pgml.train('Breast Cancer Detection', algorithm => 'random_forest'
 
 -- Gradient Boosting
 SELECT * FROM pgml.train('Breast Cancer Detection', algorithm => 'xgboost', hyperparams => '{"n_estimators": 10}');
-SELECT * FROM pgml.train('Breast Cancer Detection', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
+-- SELECT * FROM pgml.train('Breast Cancer Detection', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
 -- SELECT * FROM pgml.train('Breast Cancer Detection', algorithm => 'lightgbm', hyperparams => '{"n_estimators": 1}');
 -- Histogram Gradient Boosting is too expensive for normal tests on even a toy dataset
 -- SELECT * FROM pgml.train('Breast Cancer Detection', algorithm => 'hist_gradient_boosting', hyperparams => '{"max_iter": 2}');
diff --git a/pgml-extension/examples/image_classification.sql b/pgml-extension/examples/image_classification.sql
@@ -60,7 +60,7 @@ SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'random_forest', hyp
 
 -- gradient boosting
 SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'xgboost', hyperparams => '{"n_estimators": 10}');
-SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
+-- SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
 -- SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'lightgbm', hyperparams => '{"n_estimators": 1}');
 -- Histogram Gradient Boosting is too expensive for normal tests on even a toy dataset
 -- SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'hist_gradient_boosting', hyperparams => '{"max_iter": 2}');
diff --git a/pgml-extension/examples/joint_regression.sql b/pgml-extension/examples/joint_regression.sql
@@ -56,7 +56,7 @@ SELECT * FROM pgml.train_joint('Exercise vs Physiology', algorithm => 'random_fo
 
 -- gradient boosting
 SELECT * FROM pgml.train_joint('Exercise vs Physiology', algorithm => 'xgboost', hyperparams => '{"n_estimators": 10}');
-SELECT * FROM pgml.train_joint('Exercise vs Physiology', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
+--SELECT * FROM pgml.train_joint('Exercise vs Physiology', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
 -- SELECT * FROM pgml.train_joint('Exercise vs Physiology', algorithm => 'lightgbm', hyperparams => '{"n_estimators": 1}');
 -- Histogram Gradient Boosting is too expensive for normal tests on even a toy dataset
 -- SELECT * FROM pgml.train_joint('Exercise vs Physiology', algorithm => 'hist_gradient_boosting', hyperparams => '{"max_iter": 10}');
diff --git a/pgml-extension/examples/multi_classification.sql b/pgml-extension/examples/multi_classification.sql
@@ -53,7 +53,7 @@ SELECT * FROM pgml.train('Iris Flower Types', algorithm => 'random_forest', hype
 
 -- gradient boosting
 SELECT * FROM pgml.train('Iris Flower Types', algorithm => 'xgboost', hyperparams => '{"n_estimators": 10}');
-SELECT * FROM pgml.train('Iris Flower Types', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
+--SELECT * FROM pgml.train('Iris Flower Types', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
 -- SELECT * FROM pgml.train('Iris Flower Types', algorithm => 'lightgbm', hyperparams => '{"n_estimators": 1}');
 -- Histogram Gradient Boosting is too expensive for normal tests on even a toy dataset
 -- SELECT * FROM pgml.train('Iris Flower Types', algorithm => 'hist_gradient_boosting', hyperparams => '{"max_iter": 2}');
diff --git a/pgml-extension/examples/regression.sql b/pgml-extension/examples/regression.sql
@@ -80,7 +80,7 @@ SELECT * FROM pgml.train('Diabetes Progression', algorithm => 'random_forest', h
 
 -- gradient boosting
 SELECT * FROM pgml.train('Diabetes Progression', algorithm => 'xgboost', hyperparams => '{"n_estimators": 10}');
-SELECT * FROM pgml.train('Diabetes Progression', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
+--SELECT * FROM pgml.train('Diabetes Progression', algorithm => 'xgboost_random_forest', hyperparams => '{"n_estimators": 10}');
 -- SELECT * FROM pgml.train('Diabetes Progression', algorithm => 'lightgbm', hyperparams => '{"n_estimators": 1}');
 -- Histogram Gradient Boosting is too expensive for normal tests on even a toy dataset
 -- SELECT * FROM pgml.train('Diabetes Progression', algorithm => 'hist_gradient_boosting', hyperparams => '{"max_iter": 10}');
diff --git a/pgml-extension/examples/transformers.sql b/pgml-extension/examples/transformers.sql
@@ -33,13 +33,16 @@ SELECT pgml.transform(
     ]
 );
 
-SELECT pgml.load_dataset('opus_books', 'en-fr');
 SELECT pgml.load_dataset('kde4', kwargs => '{"lang1": "en", "lang2": "es"}');
+CREATE OR REPLACE VIEW kde4_en_to_es AS
+SELECT translation->>'en' AS "en", translation->>'es' AS "es"
+FROM pgml.kde4
+LIMIT 10;
 SELECT pgml.tune(
     'Translate English to Spanish',
-    task => 'translation_en_to_es',
-    relation_name => 'pgml.kde4',
-    y_column_name => 'translation',
+    task => 'translation',
+    relation_name => 'kde4_en_to_es',
+    y_column_name => 'es', -- translate into spanish
     model_name => 'Helsinki-NLP/opus-mt-en-es',
     hyperparams => '{
         "learning_rate": 2e-5,
@@ -49,7 +52,7 @@ SELECT pgml.tune(
         "weight_decay": 0.01,
         "max_length": 128
     }',
-    test_size => 0.05,
+    test_size => 0.5,
     test_sampling => 'last'
 );
 
