Fix merge conflits

code/bolsonaro/models/omp_forest.py

@@ -136,9 +136,9 @@ class SingleOmpForest(OmpForest):
         Make all the base tree predictions
 
         :param X: a Forest
-        :return: a np.array of the predictions of the entire forest
+        :return: a np.array of the predictions of the trees selected by OMP without applyong the weight
         """
-        forest_predictions = self._base_estimator_predictions(X).T
+        forest_predictions = np.array([tree.predict(X) for tree in self._base_forest_estimator.estimators_])
 
         if self._models_parameters.normalize_D:
             forest_predictions = forest_predictions.T
@@ -146,7 +146,5 @@ class SingleOmpForest(OmpForest):
             forest_predictions = forest_predictions.T
 
         weights = self._omp.coef_
-        omp_trees_indices = np.nonzero(weights)[0]
-
-        select_trees = np.mean(forest_predictions[omp_trees_indices], axis=0)
+        select_trees = np.mean(forest_predictions[weights != 0], axis=0)
         return select_trees

code/bolsonaro/models/omp_forest_classifier.py

@@ -42,9 +42,7 @@ class OmpForestBinaryClassifier(SingleOmpForest):
             forest_predictions = forest_predictions.T
 
         weights = self._omp.coef_
-        omp_trees_indices = np.nonzero(weights)
-
-        omp_trees_predictions = forest_predictions[omp_trees_indices].T[1]
+        omp_trees_predictions = forest_predictions[weights != 0].T[1]
 
         # Here forest_pred is the probability of being class 1.
 

code/compute_results.py

@@ -366,7 +366,7 @@ if __name__ == "__main__":
             omp_with_params_experiment_score_metric = extract_scores_across_seeds_and_extracted_forest_sizes(
                 args.models_dir, args.results_dir, int(args.experiment_ids[2]))
         #omp_with_params_without_weights
-        logger.info('Loading omp_with_params experiment scores...')
+        logger.info('Loading omp_no_weights experiment scores...')
         omp_with_params_without_weights_train_scores, omp_with_params_without_weights_dev_scores, omp_with_params_without_weights_test_scores, _, \
             omp_with_params_experiment_score_metric = extract_scores_across_seeds_and_extracted_forest_sizes(
                 args.models_dir, args.results_dir, int(args.experiment_ids[2]), weights=False)

scripts/run_compute_results.sh

 seeds='1 2 3'
-for dataset in california_housing #kin8nm kr-vs-kp spambase steel-plates diabetes diamonds boston california_housing #lfw_pairs diamonds boston iris diabetes digits linnerud wine breast_cancer olivetti_faces 20newsgroups_vectorized california_housing
+for dataset in kin8nm kr-vs-kp spambase steel-plates california_housing boston iris diabetes digits wine breast_cancer olivetti_faces diamonds
 do
+    python code/compute_results.py --stage=1 --experiment_ids 1 2 3 4 5 6 --dataset_name=$dataset --models_dir=models/$dataset/stage1
+    python code/compute_results.py --stage=2 --experiment_ids 1 2 3 4 --dataset_name=$dataset --models_dir=models/$dataset/stage2
+    python code/compute_results.py --stage=3 --experiment_ids 1 2 3 --dataset_name=$dataset --models_dir=models/$dataset/stage3
     python code/compute_results.py --stage=4 --experiment_ids 1 2 3 --dataset_name=$dataset --models_dir=models/$dataset/stage4
-    #python code/compute_results.py --stage=5 --experiment_ids 1 2 3 kmeans=5 --dataset_name=$dataset --models_dir=models/$dataset/stage5
-    #python code/compute_results.py --stage=5 --experiment_ids 1 2 3 ensemble=5 --dataset_name=$dataset --models_dir=models/$dataset/stage5_similarity
+    python code/compute_results.py --stage=5 --experiment_ids 1 2 3 similarity=4 kmeans=5 ensemble=6 --dataset_name=$dataset --models_dir=models/$dataset/stage5
 done

scripts/run_stage1_experiments.sh

@@ -5,10 +5,10 @@ seeds='1 2 3 4 5'
 
 for dataset in kin8nm kr-vs-kp spambase steel-plates california_housing boston iris diabetes digits wine breast_cancer olivetti_faces diamonds
 do
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 none_with_params --extraction_strategy=none --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=1 --models_dir=models/$dataset/stage1"
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 random_with_params --extraction_strategy=random --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=2 --models_dir=models/$dataset/stage1"
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 omp_with_params --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=3 --models_dir=models/$dataset/stage1"
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 none_wo_params --extraction_strategy=none --skip_best_hyperparams --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=4 --models_dir=models/$dataset/stage1"
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 random_wo_params --extraction_strategy=random --skip_best_hyperparams --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=5 --models_dir=models/$dataset/stage1"
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 omp_wo_params --skip_best_hyperparams --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=6 --models_dir=models/$dataset/stage1"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 none_with_params --extraction_strategy=none --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=1 --models_dir=models/$dataset/stage1"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 random_with_params --extraction_strategy=random --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=2 --models_dir=models/$dataset/stage1"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 omp_with_params --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=3 --models_dir=models/$dataset/stage1"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 none_wo_params --extraction_strategy=none --skip_best_hyperparams --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=4 --models_dir=models/$dataset/stage1"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 random_wo_params --extraction_strategy=random --skip_best_hyperparams --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=5 --models_dir=models/$dataset/stage1"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 1 omp_wo_params --skip_best_hyperparams --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --experiment_id=6 --models_dir=models/$dataset/stage1"
 done

scripts/run_stage3_experiments.sh

 #!/bin/bash
 core_number=5
-walltime=1:00
+walltime=$walltime
 seeds='1 2 3 4 5'
 
 for dataset in kin8nm kr-vs-kp spambase steel-plates california_housing boston iris diabetes digits wine breast_cancer olivetti_faces diamonds
 do
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 3 train-dev_subset --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --subsets_used=train,dev --experiment_id=1 --models_dir=models/$dataset/stage3"
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 3 train-dev_train-dev_subset --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --subsets_used=train+dev,train+dev --experiment_id=2 --models_dir=models/$dataset/stage3"
-    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=1:00 "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 3 train-train-dev_subset --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --subsets_used=train,train+dev --experiment_id=3 --models_dir=models/$dataset/stage3"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 3 train-dev_subset --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --subsets_used=train,dev --experiment_id=1 --models_dir=models/$dataset/stage3"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 3 train-dev_train-dev_subset --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --subsets_used=train+dev,train+dev --experiment_id=2 --models_dir=models/$dataset/stage3"
+    oarsub -p "(gpu is null)" -l /core=$core_number,walltime=$walltime "conda activate test_env && python code/train.py --dataset_name=$dataset --seeds $seeds --save_experiment_configuration 3 train-train-dev_subset --extracted_forest_size_stop=1 --extracted_forest_size_samples=30 --subsets_used=train,train+dev --experiment_id=3 --models_dir=models/$dataset/stage3"
 done