def download(self, path="files/names/min.csv"):
du = DameUtils()
namsorjson = path
namsorjson = open("files/names/namsor" + du.path2file(path) + ".json",
"w+")
surnames = True
names = self.csv2names(path, surnames=surnames)
namsorjson.write("[")
length = len(names)
i = 0
while (i < length):
name = names[i][0]
namsorjson.write('{"name":"' + str(names[i][0]) + '",\n')
surname = names[i][1]
namsorjson.write('"surname":"' + str(names[i][1]) + '",\n')
dnget = self.get(name=name, surname=surname, binary=True)
namsorjson.write('"gender":"' + str(dnget[0]) + '",\n')
namsorjson.write('"scale":' + str(dnget[1]) + '\n')
if ((length - 1) == i):
namsorjson.write('} \n')
else:
namsorjson.write('}, \n')
i = i + 1
namsorjson.write("]")
namsorjson.close()
def download(self, path='files/names/partial.csv', surnames=False):
du = DameUtils()
new = []
d = ""
lresult = []
res = ""
if (surnames == True):
l = self.csv2names(path, surnames=True)
for i in range(0, len(l)):
d = self.get(l[i][0], surname=l[i][1])
d["surname"] = l[i][1]
lresult.append(d)
res = str(lresult)
else:
l = self.csv2names(path)
# We must split the list in different lists with size 10
for i in range(0, len(l), 10):
new.append(l[i:i + 10])
for j in new:
lresult.append(self.get2to10(j))
for k in lresult:
res = res + str(k)
res = str(res).replace("\'", "\"")
res = str(res).replace('None', '"unknown"')
backup = open("files/names/genderize" + du.path2file(path) + ".json",
"w+")
backup.write(res)
backup.close()
return res
def test_dame_namsor_download(self):
dn = DameNamsor()
du = DameUtils()
path1 = "files/names/min.csv"
if (dn.config['DEFAULT']['namsor'] == 'yes'):
g = dn.download(path1)
self.assertTrue(
os.path.isfile("files/names/namsor" + du.path2file(path1) +
".json"))
def test_dame_genderapi_download(self):
dga = DameGenderApi()
du = DameUtils()
path1 = "files/names/min.csv"
if (dga.config['DEFAULT']['genderapi'] == 'yes'):
g = dga.download(path1)
self.assertTrue(
os.path.isfile("files/names/genderapi" + du.path2file(path1) +
".json"))
def test_path2file(self):
du = DameUtils()
self.assertEqual(du.path2file("files/images/lalla.csv"),
"files_images_lalla.csv")
parser.add_argument('--csv')
show_parser = parser.add_mutually_exclusive_group(required=False)
show_parser.add_argument('--show', dest='show', action='store_true')
show_parser.add_argument('--no-show', dest='show', action='store_false')
parser.set_defaults(show=True)
args = parser.parse_args()
du = DameUtils()
if (len(sys.argv) > 1):
#filepath = 'files/features_list.csv' #your path here
data = np.genfromtxt(args.csv, delimiter=',', dtype='float64')
scaler = MinMaxScaler(feature_range=[0, 1])
data_rescaled = scaler.fit_transform(data[1:, 0:8])
#Fitting the PCA algorithm with our Data
pca = PCA().fit(data_rescaled)
#Plotting the Cumulative Summation of the Explained Variance
plt.figure()
plt.plot(np.cumsum(pca.explained_variance_ratio_))
plt.xlabel('Number of Components')
plt.ylabel('Variance (%)') #for each component
plt.title('Dataset Explained Variance')
plt.savefig('files/images/pca_components_'+ str(du.path2file(args.csv)) + '.png')
if (args.show):
plt.show()
else:
print("You must introduce a csv file.")
print("Try $ python3 pca-components.py --csv='files/features_list.csv'")
