def get(self,link):
page_list = get_cache(link,db.GqlQuery("select * from Page where url='"+link+"'"))
page_list = list(page_list)
url_ob = None
version = self.request.get('v')
if version:
version = int(version)
if len(page_list) > 0:
url_ob = page_list[0]
if url_ob:
if not version:
version = -1
if type(version) == int and version < len(url_ob.version):
if link == '/':
url = 'Main'
else:
url = link[1:]
user_base = self.request.cookies.get('user_id')
if user_base:
user_base = user_base.split('|')[0]
else:
user_base = ''
banco = get_indices(link)[0]
tarjeta = get_indices(link)[1]
self.render('children.html',page=url_ob.page[version],url=url,page_url=all_url()[0],user_base=user_base,link=link,credito=tarjeta,banco=banco,url_full=self.request.url)
else:
self.redirect('/_edit'+link+'?v='+str(len(url_ob.version)-1))
else:
self.redirect('/_edit'+link)
def process_templates():
"""Process template"""
root = os.path.dirname(os.path.abspath(__file__))
templates_dir = os.path.join(root, 'templates')
env = Environment(loader=FileSystemLoader(templates_dir))
template = env.get_template('delete_indices_action.yml.j2')
if not os.path.exists('yml'):
os.makedirs('yml')
for i in indices.get_indices('indices.csv'):
filename = os.path.join(root, 'yml',
'delete_indices_action_' + i['index'] + '.yml')
with open(filename, 'w') as temp:
temp.write(
template.render(
index_prefix=i['index'], older_than_days=i['age']))
### TEST
start_ts = time.time()
for i in range(training_len,data.shape[0],blocksize):
if alg=='sdo':
scores[i:(i+blocksize)] = detector.predict(data[i:(i+blocksize),:])
else:
scores[i:(i+blocksize)] = detector.fit_predict(data[i:(i+blocksize),:])
print(".", end='', flush=True)
if ((i / blocksize) % 30 == 0):
print("Datapoints: ", i)
end_ts = time.time()
elapsed_ts = end_ts - start_ts
scores_file = '%sSCORES_%s_%d_T%d_%d.pickle.gz' % (outpath, alg, int(dataset_idx), int(timepam), int(idx))
with gzip.open(scores_file, 'wb') as f:
pickle.dump(scores, f)
if np.any(np.isnan(scores)):
print ('Warning: marking NaN as non-outlier')
scores[np.isnan(scores)] = 0
perf = indices.get_indices(labels[training_len:], transform_scores(scores[training_len:]))
new_row = "dataset: %d, idx: %d, P@n: %.2f, aP@n: %.2f, AP: %.2f, aAP: %.2f, MF1: %.2f, aMF1: %.2f, ROC: %.2f, Ttr: %.3f, Tts: %.3f" % (idf,idx,perf['Patn'],perf['adj_Patn'],perf['ap'],perf['adj_ap'],perf['maxf1'],perf['adj_maxf1'],perf['auc'],elapsed_tr,elapsed_ts)
res.append(new_row)
print("\n",new_row)
df = pd.DataFrame(data=res)
outfile = outpath + "SUMMARY_" + alg + "_T" + str(timepam) + ".txt"
print("Summary file (txt):",outfile,"\n")
df.to_csv(outfile, sep=',', header=None)
def post(self):
user_base = self.request.cookies.get('user_id')
if user_base:
user_base = user_base.split('|')[0]
else:
user_base = ''
banco = self.request.get('banco')
tarjeta = self.request.get('tarjeta')
datos = get_cache(banco+tarjeta+'_datos',datos_tarjeta('Promerica',self.request.get('tarjeta')))
contenido=get_cache(banco+tarjeta+'_contenido',obtener_tarjeta_promerica(self.request.get('tarjeta'),generar_promerica('Promerica')))
beneficios=get_cache(banco+tarjeta+'_beneficios',obtener_beneficios_promerica(datos[0],datos[1],datos[2],['h2','collapseomatic',get_indices(self.request.get('link'))[1]]))
content = formato_general(self.request.get('title'),contenido,beneficios)
self.render('generador.html',beneficios=beneficios,cont=contenido,contenido=content,user_base=user_base,url='Generador Promerica',link=self.request.get('link'))
def post(self):
user_base = self.request.cookies.get('user_id')
if user_base:
user_base = user_base.split('|')[0]
else:
user_base = ''
banco = self.request.get('banco')
tarjeta = self.request.get('tarjeta')
datos = get_cache(banco+tarjeta+'_datos',datos_tarjeta('LopezDeHaro',self.request.get('tarjeta')))#CONTINUAR AQUI
contenido=get_cache(banco+tarjeta+'_contenido',obtener_tarjeta_lopezdeharo(self.request.get('tarjeta'),generar_info_lopezdeharo(generar_lopezdeharo('LopezDeHaro'))))
beneficios=get_cache(banco+tarjeta+'_beneficios',obtener_beneficios(datos[0],datos[1],datos[2],['div','subtitulo',get_indices(self.request.get('link'))[1]]))
content = formato_general(self.request.get('title'),contenido,beneficios)
self.render('generador.html',beneficios=beneficios,cont=contenido,contenido=content,user_base=user_base,url='Generador LopezDeHaro',link=self.request.get('link'))
)
for k, scenario in enumerate(scenarios):
for j, algorithm in enumerate(algorithms):
for i in range(0, number_of_datasets):
datasetname = path2data + "/outlierResult_" + algorithm + "_" + scenario + "_data_" + str(
i + 1) + ".txt"
data_i = genfromtxt(datasetname, delimiter=',')
rank = data_i[train_samples:, 1].astype(float)
data_i = data_i[train_samples:, :].astype(int)
data_i[:, 2] = (data_i[:, 2] > 0).choose(data_i[:, 2], 1)
pred = np.array(data_i[:, 2], dtype=bool)
label = np.array(data_i[:, 3], dtype=bool)
label = np.invert(label)
pred = np.invert(pred)
ki = i + k * number_of_datasets
res = indices.get_indices(label, rank)
resPatn[j][ki] = res['Patn']
resap[j][ki] = res['ap']
resMF1[j][ki] = res['maxf1']
resAuc[j][ki] = res['auc']
print(" %d, %s, %d, %s, %d: %f, %f, %f, %f, %f, %f, %f, %f " %
(k, scenario, j, algorithm, i,
round(np.nanmean(resMF1[j][ki - i:ki]),
3), round(np.nanstd(resMF1[j][ki - i:ki]),
3), round(np.nanmean(resPatn[j][ki - i:ki]), 3),
round(np.nanstd(resPatn[j][ki - i:ki]),
3), round(np.nanmean(resap[j][ki - i:ki]),
3), round(np.nanstd(resap[j][ki - i:ki]), 3),
round(np.nanmean(resAuc[j][ki - i:ki]),
3), round(np.nanstd(resAuc[j][ki - i:ki]), 3)))