def train():
img_preprocess();
print ("Training Start")
obj=training(datadir,modeldir,classifier_filename)
get_file=obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
print("All Done")
def train(self):
#data_preprocess
input_datadir = './train_img'
output_datadir = './pre_img'
obj=preprocesses(input_datadir,output_datadir)
nrof_images_total,nrof_successfully_aligned=obj.collect_data()
print('Total number of images: %d' % nrof_images_total)
print('Number of successfully aligned images: %d' % nrof_successfully_aligned)
#For Training
root = Tk()
root.geometry('400x400')
root.title('Training Status')
datadir = './pre_img'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print ("Training Start")
obj=training(datadir,modeldir,classifier_filename)
get_file=obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
ourMessage = "Training Completed...!!!"
p = Label(root,text=ourMessage,fg="blue",font=("bold",20))
p.place(x=80,y=120)
root.mainloop()
def button_train(name):
scale_img = prepro.collect_data(
os.path.join(os.getcwd(), 'avengers/' + name + '/' + name + '_p.jpg'))
obj = training(modeldir, scale_img, name)
emb_array = obj.main_train()
emb_list = list(map(list, emb_array))
return jsonify({name: emb_list})
def training(descs, target):
# trnDescs = np.loadtxt('./data/trndata_input_less.csv',delimiter=',')
# trnTarget = np.loadtxt('./data/trndata_target_less.csv',delimiter=',')
trainer = classifier.training(descs, target)
trainer.set_dataset(0.15, featureNorm=True)
raw_input('Start to train?')
trainer.train_net(training_times_input=110,
num_neroun=64,
learning_rate_input=0.01,
weight_decay=0,
momentum_in=0,
verbose_input=True)
#use this network to predict all values
predictedValue = trainer.predict(trainer.network, trainer.tstdata['input'])
realValue = trainer.tstdata['class']
acc = trainer.calc_accuracy(realValue, predictedValue)
print acc #over all acc
# accuracy for each section, return a dictionary
sectionalAcc = trainer.SectionalAcc(realValue, predictedValue)
'''plot acc for each section '''
raw_input('Save?')
trainer.save_network('../net/10c_76d_Jan11.xml')
def image_train(pre_image):
datadir = './pre_img'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print("Training Start")
obj = training(modeldir, classifier_filename)
get_file = obj.main_train(pre_image)
print('Saved classifier model to file "%s"' % get_file)
def train():
datadir = './pre_img'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print("Training Start")
obj = training(datadir, modeldir, classifier_filename)
get_file = obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
sys.exit("All Done")
def button_train(name):
img = cv2.imread(
os.path.join(os.getcwd(), 'avengers/' + name + '/' + name + '_p.jpg'))
scale_img = prepro.collect_data(img)
obj = training(modeldir, scale_img, name)
emb_array = obj.main_train()
feature_list.append(feature_map(name, emb_array))
return "Success!"
def entrenarte():
#Entrenamiento
datadir = './static/pre_img'
modeldir = './modelo_transferlearning/20170511-185253.pb'
classifier_filename = './resultados/classifier2.pkl'
print("Entrenando")
obj = training(datadir, modeldir, classifier_filename)
get_file = obj.main_train()
print('Guardado en la carpeta resultados "%s"' % get_file)
# sys.exit("Termino el entrenamiento")
return render_template('entrenado.html')
def training(self):
try:
datadir = './pre_img'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print("Training Start")
obj = training(datadir, modeldir, classifier_filename)
get_file = obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
sys.exit("All Done")
return 'OK'
except Exception as error :
return error
def StartTraining():
global StopTrain
train.configure(state=tk.DISABLED)
datadir = './pre_img'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print("Training Start")
obj = training(datadir, modeldir, classifier_filename)
get_file = obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
StopTrain = True
train.configure(state=tk.ACTIVE)
text.insert(tk.INSERT, "\nCompleted...")
timer.destroy()
def button_train():
global feature_arr, feat
feature_arr = 0
print(
"========================================================================================="
)
img = request.json['face_list']
img_np = np.array(img)
obj = training(modeldir, img_np)
feature_arr = obj.main_train()
feat = True
return "Success!"
def startTraining(cwd):
os.chdir(cwd + '/FaceRecognition/')
if not hasattr(sys, 'argv'):
sys.argv = ['']
sys.path.append('.')
datadir = './pre_img'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print("Training Start")
obj = training(datadir, modeldir, classifier_filename)
get_file = obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
return "Training completed."
def click2():
"""
respond to the button2 click
"""
# toggle button colors as a test
if (button_flag[2] % 2 ==1):
button2.config(bg="white")
button_flag[2] +=1
if button_flag[2] == 1:
#pre_image.append(aug_img)
print ("Training Start")
scale_img = prepro.collect_data(os.path.join(os.getcwd(),'avengers/hermsworth/hermsworth_p.jpg'))
obj = training(modeldir, scale_img, "hermsworth")
get_feature = obj.main_train()
feature_list.append(get_feature)
print('Getting feature map succeed')
elif(button_flag[2] %2 ==0):
button2.config(bg="green")
button_flag[2] += 1
def sys_train():
try:
input_datadir = './static/people_photo/train_img'
output_datadir = './pre_img'
obj=preprocesses(input_datadir,output_datadir)
nrof_images_total,nrof_successfully_aligned=obj.collect_data()
print('Total number of images: %d' % nrof_images_total)
print('Number of successfully aligned images: %d' % nrof_successfully_aligned)
datadir = './pre_img'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print ("Training Start")
obj=training(datadir,modeldir,classifier_filename)
get_file=obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
sys.exit("All Done")
except:
print(sys.exc_info()[0])
def train(descs, target):
trainer = classifier.training(descs, target)
trainer.set_dataset(0.15, featureNorm=True)
raw_input('Start to train?')
trainer.train_net(training_times_input=200,
num_neroun=64,
learning_rate_input=0.01,
weight_decay=0.001,
momentum_in=0,
verbose_input=True)
#use this network to predict all values
predictedValue = trainer.predict(trainer.network, trainer.tstdata['input'])
realValue = trainer.tstdata['class']
acc = trainer.calc_accuracy(realValue, predictedValue)
print acc # over all acc
# accuracy for each section, return a dictionary
sectionalAcc = trainer.SectionalAcc(realValue, predictedValue)
'''plot acc for each section '''
raw_input('Save?')
trainer.save_network('./net/9c_1121chem_59d_Oct28.xml')
def train(self):
#data_preprocess
root1 = Tk()
root1.geometry('400x400')
input_datadir = './train_img'
output_datadir = './pre_img'
obj = preprocesses(input_datadir, output_datadir)
nrof_images_total, nrof_successfully_aligned = obj.collect_data()
print('Total number of images: %d' % nrof_images_total)
print('Number of successfully aligned images: %d' %
nrof_successfully_aligned)
w = Label(root1, text=str(nrof_images_total))
w1 = Label(root1, text=str(nrof_successfully_aligned))
w.pack()
w1.pack()
#For Training
root = Tk()
root.geometry('400x400')
root.title('Training Status')
datadir = './pre_img'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print("Training Start")
obj = training(datadir, modeldir, classifier_filename)
get_file = obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
ourMessage = "All Done"
p = Label(root, text=ourMessage)
p.place(x=45, y=80)
root.mainloop()
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import sys
from classifier import training
datadir = './pre_img'
modeldir = './model/20170512-110547.pb'
classifier_filename = './classifier/classifier_2017.pkl'
print("Training Start")
obj = training(datadir, modeldir, classifier_filename)
get_file = obj.main_train()
print('Saved classifier model to file "%s"' % get_file)
sys.exit("All Done")
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import sys
from classifier import training
import config
datadir = './faces_dir'
modeldir = './model/20170511-185253.pb'
classifier_filename = './class/classifier.pkl'
print('Training start')
# obj = training(datadir, modeldir, classifier_filename)
obj = training(config.FACES_DIR, config.MODEL_DIR, config.CLASSIFIER_FILENAME)
get_file = obj.main_train()
print('Saved classifier model to file "%s' % get_file)
sys.exit("All done!")
