Exemplos de FixedSplitGridEvaluator em Python

Exemplo n.º 1

import common
from dbispipeline.evaluators import FixedSplitGridEvaluator
import dbispipeline.result_handlers as result_handlers
from dbispipeline.utils import prefix_path
from loaders.librosa_features import LibRosaLoader
from sklearn.neighbors import KNeighborsClassifier
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler

dataloader = LibRosaLoader(
    prefix_path("autotagging_moodtheme-train-librosa.pickle",
                common.DEFAULT_PATH),
    prefix_path("autotagging_moodtheme-test-librosa.pickle",
                common.DEFAULT_PATH),
)

pipeline = Pipeline([("scaler", StandardScaler()),
                     ("model", KNeighborsClassifier())])

evaluator = FixedSplitGridEvaluator(
    params={
        "model__n_neighbors": [1, 3, 5, 10],
    },
    grid_params=common.grid_params(),
)

result_handlers = [
    result_handlers.print_gridsearch_results,
]

Exemplo n.º 2

from dbispipeline.evaluators import FixedSplitGridEvaluator
import dbispipeline.result_handlers as result_handlers
from dbispipeline.utils import prefix_path
from loaders.acousticbrainz import AcousticBrainzLoader
from sklearn.multioutput import MultiOutputClassifier
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC

dataloader = AcousticBrainzLoader(
    training_path=prefix_path("accousticbrainz-train.pickle",
                              common.DEFAULT_PATH),
    test_path=prefix_path("accousticbrainz-test.pickle", common.DEFAULT_PATH),
    validation_path=prefix_path("accousticbrainz-validation.pickle",
                                common.DEFAULT_PATH),
)

pipeline = Pipeline([
    ("scaler", StandardScaler()),
    ("model", MultiOutputClassifier(SVC(probability=True))),
])

evaluator = FixedSplitGridEvaluator(
    params={"model__estimator__C": [0.1, 1.0, 10.0]},
    grid_params=common.grid_params(),
)

result_handlers = [
    result_handlers.print_gridsearch_results,
]

Exemplo n.º 3

Arquivo: baseline_et.py Projeto: dbis-uibk/MediaEval2019

import dbispipeline.result_handlers as result_handlers
from dbispipeline.utils import prefix_path
from loaders.librosa_features import LibRosaLoader
from sklearn.ensemble import ExtraTreesClassifier
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler

dataloader = LibRosaLoader(
    prefix_path("autotagging_moodtheme-train-librosa.pickle",
                common.DEFAULT_PATH),
    prefix_path("autotagging_moodtheme-test-librosa.pickle",
                common.DEFAULT_PATH),
)

pipeline = Pipeline([
    ("scaler", StandardScaler()),
    ("model", ExtraTreesClassifier()),
])

evaluator = FixedSplitGridEvaluator(
    params={
        "model__n_estimators": [5, 10, 25, 100],
        "model__class_weight": [None, "balanced"],
    },
    grid_params=common.grid_params(),
)

result_handlers = [
    result_handlers.print_gridsearch_results,
]

Exemplo n.º 4

Arquivo: approach_crnn.py Projeto: dbis-uibk/MediaEval2019

dataloader = MelSpectrogramsLoader(
    data_path=prefix_path("melspec_data", common.DEFAULT_PATH),
    training_path=prefix_path("autotagging_moodtheme-train.tsv",
                              common.DEFAULT_PATH),
    test_path=prefix_path("autotagging_moodtheme-test.tsv",
                          common.DEFAULT_PATH),
    validate_path=prefix_path("autotagging_moodtheme-validation.tsv",
                              common.DEFAULT_PATH),
    window_size=WINDOW_SIZE,
)

pipeline = Pipeline([
    ("model", CRNNModel(dataloader=dataloader)),
])

grid_params = common.grid_params()
grid_params['n_jobs'] = 1

evaluator = FixedSplitGridEvaluator(
    params={
        "model__epochs": [2, 16],
        "model__output_dropout": [None],
        "model__attention": [True],
    },
    grid_params=grid_params,
)

result_handlers = [
    result_handlers.print_gridsearch_results,
]

Exemplo n.º 5

                              common.DEFAULT_PATH),
    mel_validate_path=prefix_path("autotagging_moodtheme-validation.tsv",
                                  common.DEFAULT_PATH),
    ess_training_path=prefix_path("accousticbrainz-train.pickle",
                                  common.DEFAULT_PATH),
    ess_test_path=prefix_path("accousticbrainz-test.pickle",
                              common.DEFAULT_PATH),
    ess_validate_path=prefix_path("accousticbrainz-validation.pickle",
                                  common.DEFAULT_PATH),
    window='random',
    num_windows=5,
)

pipeline = Pipeline([("model", CRNNPlusModel(dataloader=dataloader))])

grid_params = common.grid_params()
grid_params['n_jobs'] = 1

evaluator = FixedSplitGridEvaluator(
    params={
        "model__epochs": [8, 16],
        "model__output_dropout": [0.3],
        "model__concat_bn": [True],
    },
    grid_params=grid_params,
)

result_handlers = [
    result_handlers.print_gridsearch_results,
]

Exemplo n.º 6

Arquivo: baseline_neural.py Projeto: dbis-uibk/MediaEval2019

import common
from dbispipeline.evaluators import FixedSplitGridEvaluator
import dbispipeline.result_handlers as result_handlers
from dbispipeline.utils import prefix_path
from loaders.librosa_features import LibRosaLoader
from sklearn.neural_network import MLPClassifier
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler

dataloader = LibRosaLoader(
    prefix_path("autotagging_moodtheme-train-librosa.pickle",
                common.DEFAULT_PATH),
    prefix_path("autotagging_moodtheme-test-librosa.pickle",
                common.DEFAULT_PATH),
)

pipeline = Pipeline([("scaler", StandardScaler()), ("model", MLPClassifier())])

evaluator = FixedSplitGridEvaluator(
    params={},
    grid_params=common.grid_params(),
)

result_handlers = [
    result_handlers.print_gridsearch_results,
]

Exemplo n.º 7

Arquivo: config_reddit.py Projeto: bmurauer/cross_language_authorship_replication

    (f'{base}/R-CL3', 'pt', 'en'),
    (f'{base}/R-CL4', 'en', 'nl'),
    (f'{base}/R-CL4', 'nl', 'en'),
    (f'{base}/R-CL5', 'en', 'fr'),
    (f'{base}/R-CL5', 'fr', 'en'),
    (f'{base}/R-CL6', 'en', 'ar'),
    (f'{base}/R-CL6', 'ar', 'en'),
]

dataloader = MultiLoaderGenerator(RedditCrossCategoryLoader, loader_parameters)

pipeline = Pipeline([
    ("filereader", FileReader()),
    ("json", JsonTransformer()),
    ("text", DictFieldTransformer('body')),
    ('life', LifeVectorizer(samples=20, force=True)),
    ('rf', RandomForestClassifier(n_estimators=10, n_jobs=1)),
])

evaluator = FixedSplitGridEvaluator (
    {
        'life__fragment_sizes': [[50], [50, 100], [50, 100, 200]],
    }, {
        'n_jobs': -1,
        'scoring': 'accuracy',
        'verbose': 1,
    }
)

result_handlers = []

Exemplo n.º 8

Arquivo: approach_crnn_plus.py Projeto: dbis-uibk/MediaEval2019

dataloader = MelSpectrogramsLoader(
    data_path=prefix_path("melspec_data", common.DEFAULT_PATH),
    training_path=prefix_path("autotagging_moodtheme-train.tsv",
                              common.DEFAULT_PATH),
    test_path=prefix_path("autotagging_moodtheme-test.tsv",
                          common.DEFAULT_PATH),
    validate_path=prefix_path("autotagging_moodtheme-validation.tsv",
                              common.DEFAULT_PATH),
)

pipeline = Pipeline([("model",
                      CRNNPlusModel(dataloader=dataloader,
                                    essentia_loader=ab_loader))])

grid_params = common.grid_params()
grid_params['n_jobs'] = 1

evaluator = FixedSplitGridEvaluator(
    params={
        "model__epochs": [2, 4, 8, 16, 32],
        "model__crnn_output_dropout": [None],
        "model__concat_bn": [True],
    },
    grid_params=grid_params,
)

result_handlers = [
    result_handlers.print_gridsearch_results,
]