"""Config for a linear regression model evaluated on a diabetes dataset.""" from dbispipeline.evaluators import GridEvaluator import dbispipeline.result_handlers as result_handlers from sklearn.pipeline import Pipeline from sklearn.preprocessing import StandardScaler from sklearn.svm import LinearSVR from nlp4musa2020.dataloaders.alf200k import ALF200KLoader, genre_target_labels from nlp4musa2020.dataloaders.vectorizer import tfidf import nlp4musa2020.evaluators as evaluators from nlp4musa2020.models.simplenn_genre import SimpleGenreNN dataloader = ALF200KLoader('data/processed/dataset-lfm-genres.pickle', load_feature_groups=[], text_vectorizers=tfidf(), target=genre_target_labels()) pipeline = Pipeline([ ('scaler', StandardScaler()), ('model', SimpleGenreNN(epochs=50)), ]) evaluator = GridEvaluator( parameters={ 'model__dense_sizes': [ (32, 32), (64, 64), ], 'model__dropout_rate': [ 0.1, ],
"""Plan for a random forest classifier model.""" from dbispipeline.evaluators import GridEvaluator import dbispipeline.result_handlers as result_handlers from sklearn.ensemble import ExtraTreesClassifier from sklearn.pipeline import Pipeline from sklearn.preprocessing import StandardScaler from nlp4musa2020.dataloaders.alf200k import ALF200KLoader from nlp4musa2020.dataloaders.alf200k import genre_target_labels import nlp4musa2020.evaluators as evaluators dataloader = ALF200KLoader( path='data/processed/dataset-lfm-genres.pickle', load_feature_groups=[ 'statistical', ], text_vectorizers=None, target=genre_target_labels(), ) pipeline = Pipeline([ ('scaler', StandardScaler()), ('model', ExtraTreesClassifier(n_jobs=-1)), ]) evaluator = GridEvaluator( parameters={ 'model__n_estimators': [10, 100, 300], }, grid_parameters=evaluators.grid_parameters_genres(), )
import dbispipeline.result_handlers as result_handlers from sklearn.svm import LinearSVR from sklearn.pipeline import Pipeline from sklearn.preprocessing import StandardScaler from nlp4musa2020.dataloaders.vectorizer import lda, tfidf from nlp4musa2020.dataloaders.alf200k import ALF200KLoader, genre_target_labels import nlp4musa2020.evaluators as evaluators from sklearn.svm import LinearSVC from sklearn.multioutput import MultiOutputClassifier dataloader = ALF200KLoader(path='data/processed/dataset-lfm-genres.pickle', load_feature_groups=[ 'rhymes', 'statistical', 'statistical_time', 'explicitness', ], text_vectorizers=lda() + tfidf(), target=genre_target_labels()) pipeline = Pipeline([ ('scaler', StandardScaler()), ('model', MultiOutputClassifier(LinearSVC())), ]) evaluator = GridEvaluator( parameters={ 'model__estimator__C': [ 0.1, 0.5,
"""Plan for a knn model, explicit.""" from dbispipeline.evaluators import GridEvaluator import dbispipeline.result_handlers as result_handlers from sklearn.neighbors import KNeighborsClassifier from sklearn.pipeline import Pipeline from sklearn.preprocessing import StandardScaler from nlp4musa2020.dataloaders.alf200k import ALF200KLoader import nlp4musa2020.evaluators as evaluators dataloader = ALF200KLoader(path='data/processed/dataset-lfm-genres.pickle', load_feature_groups=[ 'explicitness', ], text_vectorizers=None, target=[ 'alternative', 'blues', 'country', 'dance', 'electronic', 'funk', 'hip hop', 'indie', 'jazz', 'metal', 'pop', 'punk', 'rap', 'rnb', 'rock', 'soul' ]) pipeline = Pipeline([ ('scaler', StandardScaler()), ('model', KNeighborsClassifier(n_jobs=-1, algorithm='ball_tree')), ]) evaluator = GridEvaluator( parameters={ 'model__n_neighbors': [3, 4, 5, 10], 'model__weights': ['distance'],
"""Config for a linear regression model evaluated on a diabetes dataset.""" from dbispipeline.evaluators import GridEvaluator import dbispipeline.result_handlers as result_handlers from sklearn.pipeline import Pipeline from sklearn.preprocessing import StandardScaler from sklearn.svm import LinearSVR from nlp4musa2020.dataloaders.alf200k import ALF200KLoader, genre_target_labels import nlp4musa2020.evaluators as evaluators from sklearn.svm import LinearSVC from sklearn.multioutput import MultiOutputClassifier dataloader = ALF200KLoader('data/processed/dataset-lfm-genres.pickle', load_feature_groups=[ 'explicitness', ], text_vectorizers=None, target=genre_target_labels()) pipeline = Pipeline([ ('scaler', StandardScaler()), ('model', MultiOutputClassifier(LinearSVC())), ]) evaluator = GridEvaluator( parameters={ 'model__estimator__C': [ 0.1, 0.5, 1.0, 2.0,