def get_dataset_by_features(file_name, FEATURES_NAME, mod):
try:
with open(file_name, newline='') as csvfile:
file = csv.DictReader(csvfile)
dataset = {feature_name: [] for feature_name in FEATURES_NAME}
if not file.fieldnames:
print_error("Please do not delete data from the dataset.")
elif mod == 1 and file.fieldnames[6:] != FEATURES_NAME:
print_error("Rows names of '" + file_name + "' are not valid.")
elif mod == 2 and [file.fieldnames[1]
] + file.fieldnames[6:] != FEATURES_NAME:
print_error("Rows names of '" + file_name + "' are not valid.")
for elem in file:
for feature_name in FEATURES_NAME:
if feature_name == 'Hogwarts House' and (
elem[feature_name] == 'Ravenclaw'
or elem[feature_name] == 'Slytherin'
or elem[feature_name] == 'Gryffindor'
or elem[feature_name] == 'Hufflepuff'):
dataset[feature_name].append(elem[feature_name])
else:
dataset[feature_name].append(
convert_to_number(elem[feature_name]))
return dataset
except FileNotFoundError:
print_error("Please don't modify/delete any datasets.")
def get_dataset_standardized(file_name, CONST, mod, for_standardization):
try:
with open(file_name, newline='') as csvfile:
file = csv.DictReader(csvfile)
dataset = {house_name: []
for house_name in CONST.HOUSES_NAME
} if mod == 1 else []
if not file.fieldnames:
print_error("Please do not delete data from the dataset.")
elif [file.fieldnames[1]
] + file.fieldnames[6:] != CONST.FEATURES_NAME:
print_error("Rows names of '" + file_name + "' are not valid.")
for elem in file:
new_elem = []
for valid_feature in CONST.VALID_FEATURES:
new_elem.append(
standardize(
convert_to_number(elem[valid_feature],
for_standardization,
valid_feature),
for_standardization[valid_feature]['mean'],
for_standardization[valid_feature]['std']))
dataset[elem['Hogwarts House']].append(
new_elem) if mod == 1 else dataset.append(new_elem)
return dataset
except FileNotFoundError:
print_error("Please don't modify/delete any datasets.")
def get_dataset_by_houses(file_name, FEATURES_NAME, HOUSES_NAME):
try:
with open(file_name, newline='') as csvfile:
file = csv.DictReader(csvfile)
dataset = {
house_name:
{feature_name: []
for feature_name in FEATURES_NAME[1:]}
for house_name in HOUSES_NAME
}
if not file.fieldnames:
print_error("Please do not delete data from the dataset.")
elif [file.fieldnames[1]] + file.fieldnames[6:] != FEATURES_NAME:
print_error("Rows names of '" + file_name + "' are not valid.")
for elem in file:
for feature_name in FEATURES_NAME[1:]:
dataset[elem[FEATURES_NAME[0]]][feature_name].append(
convert_to_number(elem[feature_name]))
return dataset
except FileNotFoundError:
print_error("Please don't modify/delete any datasets.")
def is_valid_dataset(dataset_name):
if dataset_name != 'datasets/dataset_test.csv' and dataset_name != 'datasets/dataset_train.csv':
print_error(
"Use one of the datasets present in the 'datasets' folder.")
def is_test_dataset(dataset_name):
if dataset_name != 'datasets/dataset_test.csv':
print_error("You have to use datasets/dataset_test.csv.")