def print_linear_regression_output(self):
data_to_print = dict()
data_to_print[self.x_axis] = self.data[self.x_axis]
data_to_print[self.y_axis] = self.data[self.y_axis]
data_to_print[constant.PREDICTION_EXPORT_COL] = self.predicted_y
TextUtils.print_dict(data_to_print)
print(f"Formula: Y = {self.slope} * X + {self.y_intercept}, slope of the line: {self.slope}, Y Intercept of the Line {self.y_intercept}")
def define_data_types(data_types, headers):
if len(data_types) == 0:
while True:
all_int = input(constant.ARE_ALL_INTEGERS)
if TextUtils.checks_yes(all_int):
return []
elif TextUtils.checks_no(all_int):
for header in headers:
while True:
header_type = input(
f"what is the type of the column {header.strip()}? (int, float, string)"
)
if TextUtils.checks_int(
header_type) or TextUtils.checks_str(
header_type) or TextUtils.checks_float(
header_type):
data_types.append(header_type.lower())
break
print(
f"please enter a valid input for {header}! (int, float, string)"
)
return data_types
print(constant.INVALID_YES_NO)
else:
return data_types
def get_valid_axis(text, data_frame):
while True:
axis = ExceptionUtils.select_int(text, 0, len(data_frame.headers) - 1)
if len(data_frame.data_types) == 0:
return axis
elif TextUtils.checks_int(
data_frame.data_types[axis]) or TextUtils.checks_float(
data_frame.data_types[axis]):
return axis
print(
f"Invalid Selection {axis}, Please select column of type Integer/Float"
)
def export_linear_regression_output_all(self):
date = datetime.now().strftime(constant.DATE_FOTMAT)
for i, header1 in enumerate(self.headers):
for j, header2 in enumerate(self.headers):
if not header1 == header2 and (len(self.data_types) == 0 or (not TextUtils.checks_str(self.data_types[i]) and not TextUtils.checks_str(self.data_types[j]))):
self.run_linear_regression(i, j)
self.export_linear_regression_output(date)
def main():
while True:
TextUtils.print_menu(constant.MENU_SELECT_DATA_SET)
selection = ExceptionUtils.select_int(
constant.MENU_SELECT_DATA_SET_INPUT, 1, 3)
if selection == 1:
print(f"Opening CSV {constant.DEFAULT_DATA_SET_FILE}")
data_frame = create_data_frame(
constant.DEFAULT_DATA_SET_FILE, ",",
constant.DEFAULT_DATA_SET_DATA_TYPES)
if data_frame:
use_data_frame(data_frame)
elif selection == 2:
path = input(constant.ENTER_PATH)
delimiter = input(constant.ENTER_DELIMITER)
print(f"Opening CSV {path}")
data_frame = create_data_frame(path, delimiter, [])
if data_frame:
use_data_frame(data_frame)
elif selection == 3:
break
def use_data_frame(data_frame):
while True:
TextUtils.print_menu(constant.MENU_USE_DATA_SET)
selection = ExceptionUtils.select_int(constant.MENU_USE_DATA_SET_INPUT,
1, 11)
if selection == 1:
TextUtils.print_dict(data_frame.data)
elif selection == 2:
data_frame.print_headings_with_type()
elif selection == 3:
data_frame.print_headings_with_type()
column = get_valid_axis(constant.GET_COLUMN, data_frame)
data_frame.print_deviation_calculations(column)
elif selection == 4:
data_frame.print_headings_with_type()
column = get_valid_axis(constant.GET_COLUMN, data_frame)
data_frame.plot_normal_distribution(column)
elif selection == 5:
data_frame.print_headings_with_type()
column = get_valid_axis(constant.GET_COLUMN, data_frame)
date = datetime.now().strftime(constant.DATE_FOTMAT)
data_frame.export_normal_distribution(column, date)
elif selection == 6:
data_frame.export_all_normal_distribution()
elif selection == 7:
run_regression(data_frame)
data_frame.print_linear_regression_output()
elif selection == 8:
run_regression(data_frame)
date = datetime.now().strftime(constant.DATE_FOTMAT)
data_frame.export_linear_regression_output(date)
elif selection == 9:
run_regression(data_frame)
data_frame.plot_linear_regression_output()
elif selection == 10:
data_frame.export_linear_regression_output_all()
elif selection == 11:
break
def process_row(self, row):
processed_row = []
if len(self.data_types) == 0:
for value in row:
if not ExceptionUtils.convert_to_int(value.strip()) is None:
processed_row.append(ExceptionUtils.convert_to_int(value.strip()))
else:
print(f"could not convert a value in the row, dropping row as data type is incorrect: int, value {value.strip()}, row: {row}")
return False
elif len(self.headers) != len(self.data_types):
print(f"The amount of data types does not match the amount of headers, dropping row {row}")
return False
else:
for i, value in enumerate(row):
if TextUtils.checks_int(self.data_types[i]) and not ExceptionUtils.convert_to_int(value.strip()) is None:
processed_row.append(ExceptionUtils.convert_to_int(value.strip()))
elif TextUtils.checks_str(self.data_types[i]):
processed_row.append(value.strip())
elif TextUtils.checks_float(self.data_types[i]) and not ExceptionUtils.convert_to_float(value.strip()) is None:
processed_row.append(ExceptionUtils.convert_to_float(value.strip()))
else:
print(f"could not convert a value in the row, dropping row as data type is incorrect: {self.data_types[i]} index of value: {i}, value {value.strip()}, row: {row}")
return False
return processed_row
def export_all_normal_distribution(self):
date = datetime.now().strftime(constant.DATE_FOTMAT)
for i, header in enumerate(self.headers):
if len(self.data_types) == 0 or not TextUtils.checks_str(self.data_types[i]):
self.export_normal_distribution(i, date)