def main():
df = utils.dataframe('dataset_train.csv')
colors = utils.colors()
df = df.drop(
columns=['Index', 'Hogwarts House', 'First Name', 'Last Name'])
df['Best Hand'] = df['Best Hand'].replace(to_replace=['Left', 'Right'],
value=[1, 2])
df['Year'] = df['Birthday'].apply(lambda x: int(x[0:4]))
df['Month'] = df['Birthday'].apply(lambda x: int(x[5:7]))
df['Day'] = df['Birthday'].apply(lambda x: int(x[8:10]))
df['Birthday'] = df['Birthday'].apply(lambda x: time.mktime(
datetime.datetime.strptime(x, "%Y-%m-%d").timetuple()))
nb_columns = 0
for column in df:
nb_columns += 1
i = 1
for column in df:
plt.subplot(4, math.ceil(nb_columns / 4 + 1), i)
i += 1
plt.hist(df[column],
bins=bin(df[column]),
color=colors[column],
label=column,
alpha=0.75,
edgecolor='black',
linewidth=0.5)
plt.legend(loc='upper right')
plt.show()
def main():
df = utils.dataframe('dataset_train.csv')
colors = utils.colors_house()
df = df.drop(columns=['Index', 'First Name', 'Last Name', 'Birthday', 'Best Hand'])
colors = [colors['Ravenclaw'], colors['Slytherin'], colors['Gryffindor'], colors['Hufflepuff']]
sns.set_palette(sns.color_palette(colors))
sns.pairplot(df, hue="Hogwarts House")
plt.show()
def main():
#matplotlib.use('webagg')
iterable = [
'Astronomy',
'Herbology',
'Ancient Runes',
'Divination',
]
plt.style.use('dark_background')
colors_house = utils.colors_house()
df = utils.dataframe('dataset_train.csv')
colors = utils.colors()
df_house = df['Hogwarts House']
df = df.drop(columns=[
'Index',
'First Name',
'Last Name',
'Arithmancy',
'Defense Against the Dark Arts',
'Muggle Studies',
'History of Magic',
'Potions',
'Care of Magical Creatures',
'Flying',
'Charms',
'Transfiguration',
])
allIterables = list(itertools.combinations(iterable, 2))
nb_columns = 0
for column in allIterables:
nb_columns += 1
i = 1
for el in allIterables:
plt.subplot(math.ceil(nb_columns / 3 + 1), 3, i)
i += 1
plt.scatter(df[el[0]],
df[el[1]],
c=df['Hogwarts House'].map(colors_house),
alpha=0.25,
label=[el[0][:10], el[1][:10]],
marker='o',
s=2)
plt.xticks([])
plt.yticks([])
plt.legend(loc='upper right', fontsize=5)
plt.show()
def main():
df2 = pd.DataFrame()
if not (len(sys.argv) == 1 + 1):
print('\033[91m' + '✘ Error: ' + '\033[0m' +
'CSV file is missing, please add his path as argument')
sys.exit()
df = utils.dataframe(sys.argv[1])
columnsNamesArr = df.columns.values
listOfColumnNames = list(columnsNamesArr)
dico_numerals = {}
for label in listOfColumnNames:
if df[label].dtypes == str or df[label].dtypes == object:
continue
dico_numerals[label] = {
'count': 0,
'mean': 0,
'std': 0,
'min': float('inf'),
'25%': 0,
'50%': 0,
'75%': 0,
'max': float('-inf'),
'total': 0
}
for label in listOfColumnNames:
if df[label].dtypes == str or df[label].dtypes == object:
continue
for index, row in df.iterrows():
if row[label] > dico_numerals[label]['max']:
dico_numerals[label]['max'] = row[label]
if row[label] < dico_numerals[label]['min']:
dico_numerals[label]['min'] = row[label]
if np.isnan(row[label]) != True:
dico_numerals[label]['count'] += 1
dico_numerals[label]['total'] += row[label]
if (dico_numerals[label]['count'] > 0):
dico_numerals[label]['mean'] = dico_numerals[label][
'total'] / dico_numerals[label]['count']
dico_numerals[label]['25%'] = utils.calc_quantile(
df[label].dropna(), 0.25)
dico_numerals[label]['50%'] = utils.calc_quantile(
df[label].dropna(), 0.5)
dico_numerals[label]['75%'] = utils.calc_quantile(
df[label].dropna(), 0.75)
dico_numerals[label]['std'] = utils.stdev(df[label].dropna())
describe(dico_numerals)
def main():
iterable = [
"Birthday",
"Best Hand",
"Arithmancy",
"Astronomy",
"Herbology",
"Defense Against the Dark Arts",
"Divination",
"Muggle Studies",
"Ancient Runes",
"History of Magic",
"Transfiguration",
"Potions",
"Care of Magical Creatures",
"Charms",
"Flying",
"Year",
"Month",
"Day"
]
df = utils.dataframe('dataset_train.csv')
colors = utils.colors()
df = df.drop(columns=['Index', 'Hogwarts House', 'First Name', 'Last Name'])
df['Best Hand'] = df['Best Hand'].replace(to_replace=['Left', 'Right'], value=[1, 2])
df['Year'] = df['Birthday'].apply(lambda x: int(x[0:4]))
df['Month'] = df['Birthday'].apply(lambda x: int(x[5:7]))
df['Day'] = df['Birthday'].apply(lambda x: int(x[8:10]))
df['Birthday'] = df['Birthday'].apply(lambda x: time.mktime(datetime.datetime.strptime(x, "%Y-%m-%d").timetuple()))
allIterables = list(itertools.combinations(iterable, 2))
nb_columns = 0
for column in allIterables:
nb_columns += 1
i = 1
for el in allIterables:
plt.subplot(7, math.ceil(nb_columns / 7 + 1), i)
i += 1
plt.scatter(df[el[0]], df[el[1]], color=utils.combine_hex_values(colors[el[0]], colors[el[1]]), alpha=0.25, label=[el[0], el[1]], s=3)
plt.xticks([])
plt.yticks([])
plt.legend(loc='upper right', fontsize=5)
plt.show()
def main():
df2 = pd.DataFrame()
if not (len(sys.argv) == 1+1):
print('\033[91m' + '✘ Error: ' + '\033[0m' + 'CSV file is missing, please add his path as argument')
sys.exit()
df = utils.dataframe(sys.argv[1])
columnsNamesArr = df.columns.values
listOfColumnNames = list(columnsNamesArr)
dico_objects = {}
for label in listOfColumnNames:
if df[label].dtypes == str or df[label].dtypes == object:
dico_objects[label] = {
'count': 0,
'unique': 0,
'top': 0,
'freq': 0,
'all': {}
}
for label in listOfColumnNames:
if df[label].dtypes == str or df[label].dtypes == object:
for index, row in df.iterrows():
if (isinstance(row[label], float) and np.isnan(row[label])):
continue
dico_objects[label]['count']+=1
if row[label] in dico_objects[label]['all']:
dico_objects[label]['all'][row[label]]+=1
else:
dico_objects[label]['all'][row[label]] = 1
for label in dico_objects:
i = 0
for k in dico_objects[label]['all']:
i+=1
if dico_objects[label]['all'][k] > dico_objects[label]['freq']:
dico_objects[label]['freq'] = dico_objects[label]['all'][k]
dico_objects[label]['top'] = k
dico_objects[label]['unique'] = i
describe(dico_objects)
def main():
b_df = utils.dataframe('dataset_train.csv')
colors = utils.colors_house()
colors = [
colors['Ravenclaw'], colors['Slytherin'], colors['Gryffindor'],
colors['Hufflepuff']
]
sns.set_palette(sns.color_palette(colors))
b_df['Year'] = b_df['Birthday'].apply(lambda x: int(x[0:4]))
b_df['Best Hand'] = b_df['Best Hand'].replace(to_replace=['Left', 'Right'],
value=[1, 2])
b_df['Year'] = b_df['Birthday'].apply(lambda x: int(x[0:4]))
b_df['Month'] = b_df['Birthday'].apply(lambda x: int(x[5:7]))
b_df['Day'] = b_df['Birthday'].apply(lambda x: int(x[8:10]))
b_df['Birthday'] = b_df['Birthday'].apply(lambda x: time.mktime(
datetime.datetime.strptime(x, "%Y-%m-%d").timetuple()))
df = pd.DataFrame()
df['Hogwarts House'] = b_df['Hogwarts House']
df['Best Hand'] = b_df['Best Hand']
df['Year'] = b_df['Year']
df['Month'] = b_df['Month']
df['Day'] = b_df['Day']
df['Birthday'] = b_df['Birthday']
df['First Name'] = b_df['First Name'].apply(lambda x: int(len(x)))
df['Last Name'] = b_df['Last Name'].apply(lambda x: int(len(x)))
df['Last Name Vowel'] = b_df['Last Name'].apply(
lambda x: int(count_vowel(x)))
df['First Name Vowel'] = b_df['First Name'].apply(
lambda x: int(count_vowel(x)))
df['Last Name Consonant'] = b_df['Last Name'].apply(
lambda x: int(len(x) - count_vowel(x)))
df['First Name Consonant'] = b_df['First Name'].apply(
lambda x: int(len(x) - count_vowel(x)))
sns.pairplot(df, hue="Hogwarts House")
plt.show()
def main():
#matplotlib.use('webagg')
'''
'Arithmancy',
'Astronomy',
'Herbology',
'Defense Against the Dark Arts',
'Divination',
'Muggle Studies',
'Ancient Runes',
'History of Magic',
'Transfiguration',
'Potions',
'Care of Magical Creatures':,
'Charms',
'Flying',
'''
iterable = [
'Astronomy',
'Herbology',
'Ancient Runes',
'Divination',
]
plt.style.use('dark_background')
colors_house = utils.colors_house()
df = utils.dataframe('dataset_train.csv')
colors = utils.colors()
df_house = df['Hogwarts House']
allIterables = list(itertools.combinations(iterable, 2))
nb_columns = 0
for column in allIterables:
nb_columns += 1
i = 1
for el in allIterables:
plt.subplot(math.ceil(nb_columns / 3 + 1), 3, i)
i += 1
plt.scatter(df[el[0]],
df[el[1]],
c=df['Hogwarts House'].map(colors_house),
alpha=0.25,
label=[el[0][:10], el[1][:10]],
marker='o',
s=2)
df_errors = df.loc[[
184, 200, 255, 339, 443, 445, 456, 504, 515, 618, 681, 704, 815,
820, 824, 915, 941, 1078, 1098, 1113, 1191, 1282, 1419, 1435, 1444,
1446, 1448, 1515, 1525, 1596
]]
plt.scatter(df_errors[el[0]],
df_errors[el[1]],
c=df_errors['Hogwarts House'].map(colors_house),
marker='x')
plt.xticks([])
plt.yticks([])
plt.legend(loc='upper right', fontsize=5)
plt.show()
print()