def popOneDayData(self): if self.md_date is None or self.md_date.shape[0] < 1: return extracts.md_extracts(np.array([]), np.array([]), np.array([]), np.array([]), np.array([]), np.array([]), np.array([])) currentDay = self.md_date[0] nextDayIndex = 0 for dayIndex in range(self.md_date.shape[0]): if currentDay != self.md_date[dayIndex]: nextDayIndex = dayIndex break if nextDayIndex == 0: nextDayIndex = self.md_date.shape[0] oneDayDate, self.md_date = np.split(self.md_date, [nextDayIndex]) oneDayTime, self.md_time = np.split(self.md_time, [nextDayIndex]) oneDayOpen, self.md_open = np.split(self.md_open, [nextDayIndex]) oneDayClose, self.md_close = np.split(self.md_close, [nextDayIndex]) oneDayHigh, self.md_high = np.split(self.md_high, [nextDayIndex]) oneDayLow, self.md_low = np.split(self.md_low, [nextDayIndex]) oneDayVolume, self.md_volume = np.split(self.md_volume, [nextDayIndex]) extract_obj = extracts.md_extracts(oneDayDate, oneDayTime, oneDayOpen, oneDayClose, oneDayHigh, oneDayLow, oneDayVolume) return extract_obj
def split(self): if self.extract_obj.md_date.shape[0] > self.threshold: input_date, output_date = np.split(self.extract_obj.md_date, [self.threshold]) input_time, output_time = np.split(self.extract_obj.md_time, [self.threshold]) input_open, output_open = np.split(self.extract_obj.md_open, [self.threshold]) input_close, output_close = np.split(self.extract_obj.md_close, [self.threshold]) input_high, output_high = np.split(self.extract_obj.md_high, [self.threshold]) input_low, output_low = np.split(self.extract_obj.md_low, [self.threshold]) input_volume, output_volume = np.split(self.extract_obj.md_volume, [self.threshold]) input_extract_obj = extracts.md_extracts(input_date, input_time, input_open, input_close, input_high, input_low, input_volume) output_extract_obj = extracts.md_extracts(output_date, output_time, output_open, output_close, output_high, output_low, output_volume) return True, input_extract_obj, output_extract_obj else: input_extract_obj = extracts.md_extracts(np.array([]), np.array( []), np.array([]), np.array([]), np.array([]), np.array([]), np.array([])) output_extract_obj = extracts.md_extracts(np.array([]), np.array( []), np.array([]), np.array([]), np.array([]), np.array([]), np.array([])) return False, input_extract_obj, output_extract_obj
def extract_file(train_file): data_file = open(train_file) data_file_header = data_file.readline() file_data = [] for line in data_file: if (len(line.strip()) > 0): file_data.append(line) data_file_header = [x for x in data_file_header.split(',') if len(x) >= 1] data_by_field = [[x for x in y.split(',') if len(x) >= 1] for y in file_data if len(y) >= 1] md_date = np.array([x[0].strip() for x in data_by_field]) md_time = np.array([x[1].strip() for x in data_by_field]) md_open = np.array([x[5].strip() for x in data_by_field]) md_close = np.array([x[6].strip() for x in data_by_field]) md_high = np.array([x[7].strip() for x in data_by_field]) md_low = np.array([x[8].strip() for x in data_by_field]) md_volume = np.array([x[9].strip() for x in data_by_field]) extracts_obj = extracts.md_extracts(md_date, md_time, md_open, md_close, md_high, md_low, md_volume) data_file.close() return extracts_obj
def setUp(self): self.md_date = np.array([ '03/06/2012', '03/06/2012', '03/06/2012', '03/06/2012', '03/06/2012', '03/06/2012' ]) self.md_time = np.array([ '09:15:00 AM', '09:16:00 AM', '09:17:00 AM', '09:18:00 AM', '09:19:00 AM', '09:20:00 AM' ]) self.md_open = np.array( ['21119', '21119', '21126', '22119', '22119', '22126']) self.md_high = np.array( ['21130', '21127', '21126', '22130', '22127', '22126']) self.md_low = np.array( ['21101', '21111', '21105', '22101', '22111', '22105']) self.md_close = np.array( ['21115', '21125', '21108', '22115', '22125', '22108']) self.md_volume = np.array( ['662', '243', '392', '2662', '2243', '2392']) self.extract_obj = extracts.md_extracts(self.md_date, self.md_time, self.md_open, self.md_close, self.md_high, self.md_low, self.md_volume)