def load_featured_data(self):
# Check if the file for csv
self.labeled_xy = pd.read_csv(personalLibrary.find_data_filename(PipelineManager.FULL_TRAIN_DATA_PATH))
self.x_test = pd.read_csv(personalLibrary.find_data_filename(PipelineManager.FULL_TEST_DATA_PATH))
self.num_features = self.labeled_xy.shape[1] - 1
self.num_examples = self.labeled_xy.shape[0]
self.split_train_csv()
def load_featured_data(self):
# Check if the file for csv
self.labeled_xy = pd.read_csv(
personalLibrary.find_data_filename(
PipelineManager.FULL_TRAIN_DATA_PATH))
self.x_test = pd.read_csv(
personalLibrary.find_data_filename(
PipelineManager.FULL_TEST_DATA_PATH))
self.num_features = self.labeled_xy.shape[1] - 1
self.num_examples = self.labeled_xy.shape[0]
self.split_train_csv()
def write_submission(self):
self.compile_report()
self.sample_submission = pd.read_csv(personalLibrary.find_data_filename(PipelineManager.FULL_SAMPLE_SUBMISSION_PATH))
# Then see how many headers it has
columns = self.sample_submission.columns.values.tolist()
pd.DataFrame({columns[self.pred_index]: self.y_test}).to_csv(self.get_submission_path(), index=False, header=self.header)
self.FULL_SUBMISSION_PATH = self.get_submission_path()
def write_submission(self):
self.sample_submission = pd.read_csv(
personalLibrary.find_data_filename(
PipelineManager.FULL_SAMPLE_SUBMISSION_PATH))
# Then see how many headers it has
columns = self.sample_submission.columns.values.tolist()
pd.DataFrame({
columns[0]: self.y_test
}).to_csv(self.get_submission_path(), index=False, header=self.header)
self.FULL_SUBMISSION_PATH = self.get_submission_path()
