def telekom_mobil_cartele(scraper_url: str):
response_body = get_json(scraper_url)
products_html = response_body['productsHtml']
soup = bs4.BeautifulSoup(products_html, 'html5lib')
feature_kw = OrderedDict([
('mobil_min_internat', 'min interna mobil'),
('mobil_min_nat mobil_sms_nat', 'minute sms naional'),
('mobil_date', 'trafic date'),
])
packages = []
extraoptiuni = soup.select('div.extraoptiuniTabWrapper')
for extraop in extraoptiuni:
detail_url = extraop.select_one('div.prepaidButtons a')['href']
abon_id = url_last_path(detail_url, scraper_url)
abon_name = extraop.select_one('div.prepaidTabTitle strong').text
abon_price = extraop.select_one('.prepaidPrice strong').text
features = extraop.select('.prepaidTabDescription .ptd_OfferRow')
features = [(feature.select_one('p').text,
feature.select_one('h4').text) for feature in features]
characteristics = extract_features(features, feature_kw, abon_name)
package = {
'name': abon_name.strip(),
'price': format_units(abon_price),
'scraper_id_hint': abon_id.strip(),
'characteristics': characteristics
}
packages.append(package)
return json.dumps({"packages": packages})
def orange_abon(scraper_url: str):
soup = get_soup(scraper_url)
feature_kw = OrderedDict([
('mobil_min_nat mobil_sms_nat', 'minute sms naiona'),
('mobil_min_nat', 'minute naiona'),
('mobil_sms_nat', 'sms naiona'),
('mobil_min_internat', 'minute interna'),
('mobil_date', 'trafic internet'),
])
abonamente = soup.select('.carusel-abo .item')
packages = []
for abon in abonamente:
name = abon.select_one('.minbenbox.romico h3').text
price = abon.select_one('.secbenbox .lefttext').text
features = abon.select('.minbenbox.romico .minben')
features = [(f.select_one('.descben').text.strip().lower(),
str(f.contents[0]).strip()) for f in features]
characteristics = extract_features(features, feature_kw, name)
packages.append({
'name': name.strip(),
'price': format_units(price),
'characteristics': characteristics
})
return json.dumps({"packages": packages})
def images_features(image_paths, color):
features = []
for image_path in image_paths:
image = cv2.imread(image_path)
# extract features will convert from RGB to the desired color space
image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
image_features = common.extract_features(image_rgb, color_space=color)
features.append(image_features)
return features
def find_cars(image, windows, color_space):
car_seen_windows = []
for window in windows:
start = window[0]
end = window[1]
cropped_image = image[start[1]:end[1], start[0]:end[0]]
if (cropped_image.shape[1] == cropped_image.shape[0]
and cropped_image.shape[1] != 0):
cropped_image = cv2.resize(cropped_image, (64, 64))
features = common.extract_features(cropped_image, color_space)
normalized_features = feature_scaler.transform([features])
prediction = svc_classifier.predict(normalized_features)
if (prediction == 1):
car_seen_windows.append(window)
return car_seen_windows
hist_feat = True
hog_feat = True
cars = glob.glob('vehicles/**/*.png')
noncars = glob.glob('non-vehicles/**/*.png')
print('Car images:', len(cars))
print('Non-Car images:', len(noncars))
t1 = time.time()
car_features = extract_features(cars,
cspace=color_space,
hist_feat=hist_feat,
hist_bins=hist_bins,
spatial_feat=spatial_feat,
spatial_size=spatial_size,
hog_feat=hog_feat,
orient=orient,
pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel)
notcar_features = extract_features(noncars,
cspace=color_space,
hist_feat=hist_feat,
hist_bins=hist_bins,
spatial_feat=spatial_feat,
spatial_size=spatial_size,
hog_feat=hog_feat,
orient=orient,
pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,