def load_data():
'''
load data from database and do processing
'''
rate_type = RATE_TYPE.USER_RATE.value
#rate_value = RATE_VALUE.DISLIKE
rate_value = None
page = None
items, _ = get_items(rate_type=rate_type, rate_value=rate_value, page=page)
return items
def test_tagit_all():
rate_type = None
rate_value = None
page = None
items, _ = get_items(rate_type=rate_type, rate_value=rate_value, page=page)
for item in items:
fanhao = item.fanhao
rate_type = RATE_TYPE.USER_RATE
rate_value = RATE_VALUE.DISLIKE
ItemRate.saveit(rate_type, rate_value, fanhao)
def test_get_items2():
rate_type = None
rate_value = None
page = None
items, page_info = get_items(rate_type=rate_type,
rate_value=rate_value,
page=page)
assert len(items) > 0
print(f'item count:{len(items)}')
print(
f'total_items: {page_info[0]}, total_page: {page_info[1]}, current_page: {page_info[2]}, page_size:{page_info[3]}'
)
def test_get_items():
rate_type = RATE_TYPE.USER_RATE
rate_value = RATE_VALUE.LIKE
page = None
items, page_info = get_items(rate_type=rate_type,
rate_value=rate_value,
page=page)
assert len(items) > 0
print(f'item count:{len(items)}')
print(
f'total_items: {page_info[0]}, total_page: {page_info[1]}, current_page: {page_info[2]}, page_size:{page_info[3]}'
)
def prepare_predict_data():
# get not rated data
rate_type = None
rate_value = None
page = None
unrated_items, _ = get_items(rate_type=rate_type,
rate_value=rate_value,
page=page)
#mlb = load_model(get_data_path(MODEL_FILE))
dicts = as_dict(unrated_items)
lfw = create_data(dicts)
n_samples = lfw.data.shape[0]
if n_samples < MIN_TRAIN_NUM:
raise ValueError(f'训练数据不足, 无法训练模型. 需要{MIN_TRAIN_NUM}, 当前{n_samples}')
return lfw.ids, dimension(lfw.data)
def test_download_items():
rate_type = RATE_TYPE.USER_RATE
rate_value = RATE_VALUE.LIKE
page = None
items, _ = get_items(rate_type=rate_type, rate_value=rate_value, page=page)
assert len(items) > 0
try:
for item in items:
for face in item.faces_dict:
if face.value == None:
face.value = parse_face(face.url)
face = Face.updateit(face)
print('update face')
except Exception as e:
print('system error')
traceback.print_exc()
def index():
rate_type = RATE_TYPE.SYSTEM_RATE.value
rate_value = int(request.query.get('like', RATE_VALUE.LIKE.value))
page = int(request.query.get('page', 1))
items, page_info = get_items(rate_type=rate_type,
rate_value=rate_value,
page=page)
for item in items:
_remove_extra_tags(item)
today_update_count = db.get_today_update_count()
today_recommend_count = db.get_today_recommend_count()
msg = f'今日更新 {today_update_count} , 今日推荐 {today_recommend_count}'
return template('index',
items=items,
page_info=page_info,
like=rate_value,
path=request.path,
msg=msg)
def tagit():
rate_value = request.query.get('like', None)
rate_value = None if rate_value == 'None' else rate_value
rate_type = None
if rate_value:
rate_value = int(rate_value)
rate_type = RATE_TYPE.USER_RATE
page = int(request.query.get('page', 1))
items, page_info = get_items(rate_type=rate_type,
rate_value=rate_value,
page=page)
for item in items:
_remove_extra_tags(item)
return template('tagit',
items=items,
page_info=page_info,
like=rate_value,
path=request.path)
def test_download_face():
rate_type = RATE_TYPE.USER_RATE
rate_value = RATE_VALUE.LIKE
page = None
items, _ = get_items(rate_type=rate_type, rate_value=rate_value, page=page)
assert len(items) > 0
item = items[0]
face = item.faces_dict[0]
face_url = face.url
inputImg = url_to_image(face_url)
h, w = inputImg.shape[:2]
scale = 1
if h > 600 or w > 800:
scale = 600 / max(h, w)
dims = (int(w * scale), int(h * scale))
interpln = cv2.INTER_LINEAR if scale > 1.0 else cv2.INTER_AREA
inputImg = cv2.resize(inputImg, dims, interpolation=interpln)
faces = fd.detect_faces_dnn(inputImg)
faces