def main():
if len(argv) > 1:
df_raw = get_dataset(argv[1])
else:
return print(
'Please input the path to the dataset as the first argument.')
if df_raw is None:
return print('Please input a valid path to the dataset.')
possible_columnname = [
'Hogwarts House', 'Astronomy', 'Herbology', 'Divination',
'Muggle Studies', 'Ancient Runes', 'History of Magic',
'Transfiguration', 'Potions', 'Charms', 'Flying'
]
df = df_raw[possible_columnname]
df = df.dropna()
plot_pairplot(df)
return
def main():
df_raw = get_dataset('datasets/dataset_train.csv')
if df_raw is None:
return print('Please input a valid file as first argument.')
df_filter = filter_dataframe(df_raw)
df_norm = normalize_dataframe(df_filter)
possible_columnname = [
'Arithmancy', 'Astronomy', 'Herbology',
'Defense Against the Dark Arts', 'Divination', 'Muggle Studies',
'Ancient Runes', 'History of Magic', 'Transfiguration', 'Potions',
'Care of Magical Creatures', 'Charms', 'Flying'
]
if len(argv) > 1:
if argv[1] in possible_columnname:
plot_histogram(df_raw, df_norm, argv[1])
return
plot_histogram(df_raw, df_norm)
return
def main():
if len(argv) > 1:
df_raw = get_dataset(argv[1])
else:
return print(
'Please input the path to the dataset as the first argument.')
if df_raw is None:
return print('Please input a valid path to the dataset.')
df_filter = filter_dataframe(df_raw)
possible_columnname = [
'Arithmancy', 'Astronomy', 'Herbology',
'Defense Against the Dark Arts', 'Divination', 'Muggle Studies',
'Ancient Runes', 'History of Magic', 'Transfiguration', 'Potions',
'Care of Magical Creatures', 'Charms', 'Flying'
]
if len(argv) == 4:
if argv[2] in possible_columnname and argv[3] in possible_columnname:
plot_scatter(df_filter, argv[2], argv[3])
return
plot_scatter(df_filter, 'Astronomy', 'Defense Against the Dark Arts')
return
return test_houses
def arguments_parsing():
parser = argparse.ArgumentParser()
parser.add_argument('test', type=str, help='test file')
parser.add_argument('weights', type=str, help='weights file')
args = parser.parse_args()
return args
if __name__ == '__main__':
# Argument
args = arguments_parsing()
dataset, num_categories = get_dataset(args.test)
num_categories.remove('Index')
nb = len(num_categories)
dataset, useless = select_right_data_to_list(dataset, num_categories)
# Read weights
try:
with open(args.weights, 'r') as f:
w = np.zeros((4, nb + 1))
fl = f.readlines()
for i, x in enumerate(fl):
for j, n in enumerate(x.split(',')):
n = float(n) if is_number(n) else 0
w[j][i] = n
except:
w = np.zeros((4, nb + 1))
# Argument
parser = argparse.ArgumentParser()
parser.add_argument('file', type=str, help='file described')
parser.add_argument('-lf',
'--search',
help='all plot',
action="store_true")
parser.add_argument('-o', '--output', type=str, help='save file')
args = parser.parse_args()
houses = ['Gryffindor', 'Hufflepuff', 'Ravenclaw', 'Slytherin']
colors = ['red', 'yellow', 'blue', 'green']
labels = ['Gr', 'Huf', 'Rav', 'Sly']
dataset, num_categories = get_dataset(args.file)
num_categories.remove('Index')
p = 2
if not args.search:
num_categories = ['Care of Magical Creatures']
p = 1
plt.figure(figsize=(15, 20))
plt.subplots_adjust(top=0.98, wspace=0.2, hspace=0.5)
for j, c in enumerate(num_categories):
plt.subplot(math.ceil(len(num_categories) / 2), p, j + 1)
for i, h in enumerate(houses):
tmp_dataset = [
data[c] for data in dataset
if c in data and data['Hogwarts House'] == h
]