def main():
LANDMARKS_MODEL_URL = 'http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2'
face_list = cp.run(LANDMARKS_MODEL_URL)
file_list = os.listdir("./pytorch_stylegan_encoder/aligned_images")
print(file_list)
name_list = []
for name in file_list:
name_list.append(name)
#npyfile_list = os.listdir("./npysave/")
#print(npyfile_list)
#npy_list = []
#for name in npyfile_list:
# npy_list.append(name)
for i in range(len(name_list)) :
print(i," ",len(name_list))
args.image_path = "./pytorch_stylegan_encoder/aligned_images/"+name_list[i]
print(args.image_path)
args.dlatent_path="./npysave/unknown"+str(i+1)+".npy"
print("args.dlatent_path",args.dlatent_path)
optimize_latents()
main_edit(args.dlatent_path, i)
def handle(msg):
content_type, chat_type, chat_id = telepot.glance(msg)
if content_type == 'text':
text = msg['text'].upper().strip()
if text.startswith('/START') :
bot.sendMessage(chat_id, '*Digite:*\n /comandos - para ver comandos' , parse_mode='Markdown')
if text.startswith('/COMANDOS'):
comandos = "🛠*Lista de Comandos*\n\n• *FOTOS:* /fotos baixar imagens\n\n• *IP:* /find - Consultar ip\n\n• *BIN:* /bin - consultar uma bin\n\n• *TR:* /tr traduzir um texto"
bot.sendMessage(chat_id, comandos, parse_mode= 'Markdown')
if text.startswith('/MEME'):
image = PegarPost()
bot.sendPhoto(chat_id, image)
if text.startswith('/FIND'):
try:
ip = text[6:]
if ip and '@DARKMINE' not in text:
locap = locapIp(ip)
bot.sendMessage(chat_id, locap[0], parse_mode= 'Markdown')
bot.sendPhoto(chat_id, locap[1], '@Darkmine_bot')
else:
data = '*Ip Location* - Localizar IP ou Site\n\nFormato:\n/ip ip ou hostname'
bot.sendMessage(chat_id, data, parse_mode= 'Markdown')
except Exception as erro:
print(erro)
data = '*Endereço de IP inválido!*'
bot.sendMessage(chat_id, data, parse_mode= 'Markdown')
if text.startswith('/BIN'):
resp = BinChecker(text)
bot.sendMessage(chat_id, resp, parse_mode= 'Markdown')
if text.startswith('/TR'):
texto = text[4:]
if texto and '@DARKMINE' not in text:
resp = traduzirTexto(texto)
else:
resp = "*Modo de uso:*\n\n /tr *Texto* - Para fazer a tradução de algum texto"
bot.sendMessage(chat_id, resp, parse_mode= 'Markdown')
if text.startswith('/ID'):
nome = msg['from']['first_name']
user = msg['from']['username']
resp = 'Nome: {}\nUsuário: @{}\nId: {}'.format(nome, user, chat_id)
bot.sendMessage(chat_id,resp)
if text.startswith('/FOTOS'):
if text[7:] and '@DARKMINE' not in text:
text = text[7:].split(' ')
num = int(text[len(text) - 1])
query = '+'.join(text[0:len(text)-1])
lista = run(query, num)
for line in lista:
if line == lista[len(lista) - 1]:
bot.sendPhoto(chat_id, line, '@Darkmine_bot')
else:
bot.sendPhoto(chat_id, line)
else:
resp = "*Modo de uso:*\n\n/fotos *busca* *quantidade*\n\nEx: /fotos bolsonaro 5"
bot.sendMessage(chat_id, resp, parse_mode='Markdown')
data.x = F.elu(self.conv1(data.x, data.edge_index, data.edge_attr))
weight = normalized_cut_2d(data.edge_index, data.pos)
cluster = graclus(data.edge_index, weight, data.x.size(0))
data.edge_attr = None
data = max_pool(cluster, data, transform=T.Cartesian(cat=False))
data.x = F.elu(self.conv2(data.x, data.edge_index, data.edge_attr))
weight = normalized_cut_2d(data.edge_index, data.pos)
cluster = graclus(data.edge_index, weight, data.x.size(0))
data = max_pool(cluster, data, transform=T.Cartesian(cat=False))
data.x = F.elu(self.conv3(data.x, data.edge_index, data.edge_attr))
x = global_mean_pool(data.x, data.batch)
x = F.elu(self.fc1(x))
x = F.dropout(x, training=self.training)
return F.log_softmax(self.fc2(x), dim=1)
model = MoNet(args.kernel_size).to(device)
optimizer = torch.optim.Adam(model.parameters(),
lr=args.lr,
weight_decay=args.weight_decay)
scheduler = torch.optim.lr_scheduler.StepLR(optimizer,
args.decay_step,
gamma=args.lr_decay)
print(model)
run(model, args.epochs, train_loader, test_loader, optimizer, scheduler,
device)
from pathlib import Path
import cv2
from image import run
path = Path(__file__).parents[1] / 'puzzles/'
for filename in path.iterdir():
img = cv2.imread(str(filename.resolve()))
if img.shape[0] > 2000:
img = cv2.resize(img, (500, 700))
run(img)
def deconstruct_image(image):
def reconstruct_image():
return 0
tf.set_random_seed(1)
np.random.seed(1)
zzz = image.run()
# print(color_intensities)
# zzz = [ 3.45591175, 4.21306292, 1.39260529, 0.79979209, 2.80560091, 1.07702818,
# 1.13857619, 2.06239613, 1.01650569, 2.74965652, 2.89521513, 1.78206757,
# 3.03480862, 1.155502, 0.84656923, 0.94706706, 1.30763502, 14.22199015,
# 1.01494266, 0.82881535, 1.20278014, 2.6748784, 1.52933522, 0.93917355]
# gradient descent
# -1/1 8 0.1 : 0.002077
# -1/1 8 0.5 : 0.001934 then waver up and down
# -1/1 8 0.8 :
# -1/1 8 1.0 : 12.37966 .....
# adamoptimizer
# -1/1 10 0.1 : 0.022
# -1/1 10 0.001 : 0. 001126
# -1/1 10 0.0005 : 0. 0011261
# -1/1 16 0.0005 : 0.001918
# -1/1 20 0.001 : 0.001126
# -1/1 20 0.0005 : 0.000765
# -1/1 20 0.0001 : 0.0007651
# -1/1 25 0.00005 : 0.0007651
# 0/1 20 0.0001 : 0.0168
# -2/1 20 0.0001 : 0.0008762
# -0.5/0.5 20 0.0001 : 0.00109
# -2.0/2.0 20 0.0001 : 0.0016359
# -1/1 25 0.0005 : 0.0016355
# -1/1 30 0.0005 : 0.0016355
color_intensities = zzz / np.linalg.norm(zzz)
# image -> color intensity
# provided color_intensity -> avg of your color intensities
# e
# fake data
x = np.linspace(-1.0, 1.0, 25)[:, np.newaxis] # shape (100, 1)
# noise = np.random.normal(0, 0.1, size=x.shape)
y = np.asarray(color_intensities)
y.shape=(25,1) # shape (100, 1) + some noise
# plot data
# plt.scatter(x, y)
# plt.show()
tf_x = tf.placeholder(tf.float32, x.shape) # input x
tf_y = tf.placeholder(tf.float32, y.shape) # input y
# neural network layers
l1 = tf.layers.dense(tf_x, 20, tf.nn.relu) # hidden layer
output = tf.layers.dense(l1, 1) # output layer
# step = tf.Variable(0, trainable=False)
# rate = tf.train.exponential_decay(0.0005, step, 1, 0.9999)
loss = tf.losses.mean_squared_error(tf_y, output) # compute cost
optimizer = tf.train.AdamOptimizer(0.0005)
train_op = optimizer.minimize(loss)
sess = tf.Session() # control training and others
sess.run(tf.global_variables_initializer()) # initialize var in graph
plt.ioff() # something about plotting
for step in range(25000):
# train and net output
_, l, pred = sess.run([train_op, loss, output], {tf_x: x, tf_y: y})
if step % 1000 == 0:
# plot and show learning process
# plt.cla()
# plt.scatter(x, y)
# plt.plot(x, pred, 'r-', lw=5)
# plt.text(0.5, 0, 'Loss=%.4f' % l, fontdict={'size': 20, 'color': 'red'})
# plt.pause(1.0)
# plt.text(0.5, 0, 'Loss=%.4f' % l, fontdict={'size': 20, 'color': 'red'})
# plt.show()
print('Loss=%.12f' % l)
tvars = tf.trainable_variables()
tvars_vals = sess.run(tvars)
for var, val in zip(tvars, tvars_vals):
print(var.name, val) # Prints the name of the variable alongside its value.
# print(output.numpy())
plt.scatter(x, y)
plt.plot(x, pred, 'r-', lw=2)
plt.text(0.5, 0, 'Loss=%.6f' % l, fontdict={'size': 20, 'color': 'red'})
plt.pause(1.0)
plt.text(0.5, 0, 'Loss=%.6f' % l, fontdict={'size': 20, 'color': 'red'})
plt.show()
# plt.ioff()
plt.show()
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
import image
# LOAD DATA
mnist = input_data.read_data_sets('data/', one_hot=True)
images = image.run()
# INIT WEIGHTS
def init_weights(shape):
init_random_dist = tf.truncated_normal(shape, stddev=0.1)
return tf.Variable(init_random_dist)
# INIT BIAS
def init_bias(shape):
init_bias_vals = tf.constant(0.1, shape=shape)
return tf.Variable(init_bias_vals)
# CONV2D
def conv2d(x, W):
return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')
# POOLING
def max_pool_2by2(x):
return tf.nn.max_pool(x,
ksize=[1, 2, 2, 1],
strides=[1, 2, 2, 1],