def run_cv(s, fold, base_folder, params, dataset, classifier_builder):
if file_exists(base_folder + "/models/model_" + str(fold) + "_model"):
print("will not run")
return
if fold == -1:
if file_exists(base_folder + "/models/model_final_net"):
print("will not run")
return
# set random seeds
set_seeds(s)
create_dir_structure(base_folder)
# build the final model
if fold == -1:
classifier = classifier_builder(**params)
classifier.train(dataset, lc_file=base_folder + "/lcs/lc_fold_-1")
ids, predictions, classes_d = classifier.evaluate(dataset)
CrossValidation.write_results(ids, predictions, classes_d, -1,
base_folder + "/predictions/")
classifier.save_model(base_folder + "/models/model_final")
ids, projection, classes = classifier.get_projection(dataset)
cross_validation.save_projections(
projection, base_folder + "/projections/fold_train_" + str(fold),
ids)
for class_name in classes_d.columns:
ids, projection, classes = classifier.get_projection(
dataset, class_name)
cross_validation.save_projections(
projection, base_folder + "/projections/class_fold_train_" +
class_name + "_" + str(fold), ids)
else:
# run 1 fold of 10 cv, saving each model
cross_valid = CrossValidation(dataset, classifier_builder, params,
base_folder)
cross_valid.run(fold)
rna.setImputDimNeurons(30)
rna.setNumberNeuronsHiddenLayer(31)
rna.setActivationFunctionHiddenLayer("tanh")
rna.setNumberNeuronsOutputLayer(1)
rna.setActivationFunctionOutputLayer("tanh")
rna_classifier = RnaClassifier()
rna_classifier.setRna(rna)
#PREPROCESSADOR PARA ATRIBUTOS CATEGORICOS
preprocessor = Preprocessor()
preprocessor.setColumnsCategory(['protocol_type', 'service', 'flag'])
evaluate = EvaluateModule()
cross = CrossValidation()
#DEFINIR A ITERACAO QUE O CROSS VALIDATION ESTA
cross.setIteration(1)
cross.setPreprocessor(preprocessor)
cross.setFilePath("../../bases/sub_bases_nslkdd_30_attribute/")
cross.setResultPath("../../results/30_attribute/rna_oculta_31_time/")
cross.setClassifier(rna_classifier)
cross.setEvaluateModule(evaluate)
cross.run()
# from sklearn.linear_model import LogisticRegression
from sklearn.naive_bayes import MultinomialNB
import numpy as np
from feature_engineering import FeatureEngineering
from cross_validation import CrossValidation
from multi_log_loss import multi_log_loss
f = FeatureEngineering('../data/gender_age_train.csv',
'../data/gender_age_test.csv',
'device_id',
wide_files=[#'../features/apps_per_event.csv', '../features/avg_position.csv',
#'../features/count_by_hour.csv', '../features/count_by_period.csv',
'../features/event_counts.csv', '../features/sd_position.csv'],
long_files=[#'../features/active_app_category_counts.csv',
#'../features/installed_app_category_counts.csv',
'../features/phone_brand.csv'])
labels, features, colnames = f.extract_features()
labels.set_index(np.arange(labels.shape[0]), inplace=True)
colnames.set_index(np.arange(colnames.shape[0]), inplace=True)
train_filter = [i for i, x in enumerate(labels['age'].tolist()) if not np.isnan(x)]
test_filter = [i for i, x in enumerate(labels['age'].tolist()) if np.isnan(x)]
cv = CrossValidation(features[train_filter, :],
labels.ix[train_filter, 'group'],
features[test_filter, :],
multi_log_loss)
model = MultinomialNB()
model.predict = model.predict_proba
out = cv.run(model, 'test')
@staticmethod
def filter_X(X):
# pass list so a DataFrame is returned
return X.loc[:, ['age']]
def my_roc(actual, predicted):
actual = [int(x) for x in actual]
return roc_auc_score(actual, predicted)
train = pd.read_csv('gender_age_train.csv')
train['train_fl'] = True
train['gender'] = [float(x == 'M') for x in train['gender']]
test = pd.read_csv('gender_age_test.csv')
test['gender'] = ''
test['age'] = -1
test['group'] = ''
test['train_fl'] = False
data = pd.concat([train, test])
cv = CrossValidation(data,
my_roc,
'gender',
'train_fl',
id_col='device_id',
logged=True)
model = MyLasso()
cv.run(model, 'test_age_gender')
def _train(current_experiment, train_data_folder_path,
train_labels_folder_path, train_ids_path):
# this must be imported after setting CUDA_VISIBLE_DEVICES environment variable, otherwise it won't work
from cross_validation import CrossValidation
cv = CrossValidation(current_experiment)
cv.run(train_data_folder_path, train_labels_folder_path, train_ids_path)
