from sklearn.svm import LinearSVC
from sklearn.naive_bayes import GaussianNB
from sklearn.pipeline import Pipeline
from sklearn.impute import SimpleImputer
from sklearn.preprocessing import StandardScaler
from joblib import dump, load
from data_handling import load_user_data, load_some_user_data, \
split_features_labels, user_train_test_split
from metrics import balanced_accuracy_score, single_balanced_accuracy_score
from flex_one_vs_rest_classifier import FlexOneVsRestClassifier
if __name__ == "__main__":
start = time.time()
# load data and reset index
data = load_user_data("../../data")
data.reset_index(inplace=True)
X, y = split_features_labels(data)
attrs = X.columns
labels = y.columns
X = X.values
y = y.values
# save save uuid for stratified train-test-split
le = LabelEncoder()
le.fit(X[:, 0])
strat_classes = le.transform(X[:, 0])
X_train, X_test, y_train, y_test = user_train_test_split(X,
y,
test_size=0.2,
import sys
import pandas as pd
from joblib import load
from data_handling import load_user_data
if __name__ == "__main__":
args = sys.argv
data_path = args[1]
classifier_path = args[2]
output_path = args[3]
# read and process training data
data = load_user_data(data_path)
data.reset_index(inplace=True)
feature_names = data.columns
# drop uuid column, the timestamps and the label source
if "label_source" in data.columns:
data = data.drop(["label_source"], axis=1)
# drop uuid column, the timestamps and the label source
X = data.drop(['level_0', 'level_1', 'timestamp'], axis=1)
clf = load(classifier_path)
y_pred = clf.predict(X)
df = pd.DataFrame(y_pred, columns=clf.label_names)
df.to_csv(output_path)
import xgboost as xgb
import matplotlib.pyplot as plt
import pandas as pd
from sklearn.impute import SimpleImputer
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn.metrics import f1_score, confusion_matrix
from sklearn.pipeline import Pipeline
from data_handling import load_user_data, split_features_labels
if __name__ == '__main__':
# load data and reset index
user_data = load_user_data()
features_df, labels_df = split_features_labels(user_data)
del user_data
features_df.reset_index(inplace=True)
# get uuid column and remove them, the timestamps and the label_source from
# the labels
y = features_df.iloc[:, 0].values
index_cols = features_df.columns[[0, 1, 2, -1]]
features_df.drop(index_cols, axis=1, inplace=True)
feature_names = features_df.columns
X = features_df
# train-test split and wrap output in data frame to save column names
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2,
random_state=41)