def main():
model = myVGG.VGG_16()
X_fname = 'C:/Users/Sai/Desktop/Facial_Expression_Recognition/Data_Generation/data/X_train_train.npy'
y_fname = 'C:/Users/Sai/Desktop/Facial_Expression_Recognition/Data_Generation/data/y_train_train.npy'
X_train = np.load(X_fname)
y_train = np.load(y_fname)
print(X_train.shape)
print(y_train.shape)
print("Training started")
callbacks = []
earlystop_callback = EarlyStopping(monitor='val_loss', patience=5, verbose=0)
batch_print_callback = LambdaCallback(on_batch_begin=lambda batch, logs: print(batch))
epoch_print_callback = LambdaCallback(on_epoch_end=lambda epoch, logs: print("epoch:", epoch))
callbacks.append(earlystop_callback)
callbacks.append(batch_print_callback)
callbacks.append(epoch_print_callback)
batch_size = 512
model.fit(X_train, y_train, nb_epoch=400, \
batch_size=batch_size, \
validation_split=0.2, \
shuffle=True, verbose=0, \
callbacks=callbacks)
model.save_weights('C:/Users/Sai/Desktop/Facial_Expression_Recognition/CNN_Model/my_model_weights.h5')
scores = model.evaluate(X_train, y_train, verbose=0)
print ("Train loss : %.3f" % scores[0])
print ("Train accuracy : %.3f" % scores[1])
print ("Training finished")
def main():
model = myVGG.VGG_16()
if args.test is not None:
print("Test mode")
img = cv2.imread(args.test)
img = fu.preprocessing(img)
img = np.expand_dims(img, axis=0)
y = np.expand_dims(np.asarray([0]), axis=0)
batch_size = 1
model.fit(img, y, nb_epoch=400, \
batch_size=batch_size, \
validation_split=0.2, \
shuffle=True, verbose=0)
return
#input_path = args.data_path
#print("training data path : " + input_path)
#X_train, y_train = fu.extract_features(input_path)
X_fname = '../data/X_train_train.npy'
y_fname = '../data/y_train_train.npy'
X_train = np.load(X_fname)
y_train = np.load(y_fname)
print(X_train.shape)
print(y_train.shape)
print("Training started")
callbacks = []
earlystop_callback = EarlyStopping(monitor='val_loss',
patience=5,
verbose=0)
batch_print_callback = LambdaCallback(
on_batch_begin=lambda batch, logs: print(batch))
epoch_print_callback = LambdaCallback(
on_epoch_end=lambda epoch, logs: print("epoch:", epoch))
callbacks.append(earlystop_callback)
callbacks.append(batch_print_callback)
callbacks.append(epoch_print_callback)
batch_size = 512
model.fit(X_train, y_train, nb_epoch=400, \
batch_size=batch_size, \
validation_split=0.2, \
shuffle=True, verbose=0, \
callbacks=callbacks)
model.save_weights('my_model_weights.h5')
scores = model.evaluate(X_train, y_train, verbose=0)
print("Train loss : %.3f" % scores[0])
print("Train accuracy : %.3f" % scores[1])
print("Training finished")
def main():
model = myVGG.VGG_16('my_model_weights.h5')
if args.image is not None:
print('Image Prediction Mode')
img = fu.preprocessing(cv2.imread(args.image))
X = np.expand_dims(img, axis=0)
X = np.expand_dims(X, axis=0)
result = model.predict(X)
print(result)
return
elif args.dataset is not None:
print("Directory Prediction Mode")
X, y = fu.extract_features(args.dataset)
scores = model.evaluate(X, y, verbose=0)
print(scores)
return
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
os.environ['KERAS_BACKEND'] = 'tensorflow'
sys.path.append('C:/Users/Sai/Desktop/Facial_Expression_Recognition/CNN_Model')
sys.path.append(
'C:/Users/Sai/Desktop/Facial_Expression_Recognition/Model_Training')
emotion = ['Angry', 'Fear', 'Happy', 'Sad', 'Surprise', 'Neutral']
import myVGG
model = myVGG.VGG_16(
'C:/Users/Sai/Desktop/Facial_Expression_Recognition/Model_Training/my_model_weights.h5'
)
import cv2
import numpy as np
def preprocessing(img, size=(48, 48)):
img = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
img = cv2.resize(img, size).astype(np.float32)
return img
def predict_emotion(gray_face):
resized_img = cv2.resize(gray_face, (48, 48), interpolation=cv2.INTER_AREA)
image = resized_img.reshape(1, 1, 48, 48)
import myVGG
import cv2
import numpy as np
parser = argparse.ArgumentParser(description=("Testing Prediction"))
parser.add_argument('--image',
help=('Input an image to test model prediction'))
parser.add_argument('--dataset',
help=('Input a directory to test model prediction'))
parser.add_argument('--model', help=('Input a trained model'))
args = parser.parse_args()
if args.model:
model = myVGG.VGG_16(args.model)
def main():
model = myVGG.VGG_16('my_model_weights.h5')
if args.image is not None:
print('Image Prediction Mode')
img = fu.preprocessing(cv2.imread(args.image))
X = np.expand_dims(img, axis=0)
X = np.expand_dims(X, axis=0)
result = model.predict(X)
print("class distribution: ", result)
print("The best: ", np.argmax(result[0], 0))
return