def get_VQA_model(VQA_weights_file_name):
''' Given the VQA model and its weights, compiles and returns the model '''
from models.VQA.VQA import VQA_MODEL
vqa_model = VQA_MODEL()
vqa_model.load_weights(VQA_weights_file_name)
vqa_model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
return vqa_model
def get_VQA_model(VQA_weights_file_name):
''' Given the VQA model and its weights, compiles and returns the model '''
from models.VQA.VQA import VQA_MODEL
vqa_model = VQA_MODEL()
vqa_model.load_weights(VQA_weights_file_name)
vqa_model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
return vqa_model
def get_VQA_model(VQA_weights_file_name):
from models.VQA.VQA import VQA_MODEL
vqa_model = VQA_MODEL()
vqa_model.load_weights(VQA_weights_file_name)
vqa_model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
return vqa_model
def get_VQA_model(VQA_weights_file_name):
global vqa_model
vqa_model = VQA_MODEL()
vqa_model.load_weights(VQA_weights_file_name)
vqa_model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
print ("VQA Model loaded!")
def get_VQA_model(VQA_weights_file_name):
"""
Given the VQA model and its weights, compiles and returns the model
:param VQA_weights_file_name: VQA model weight file (including path)
:return: pre-trained VQA model
"""
from models.VQA.VQA import VQA_MODEL
vqa_model = VQA_MODEL()
vqa_model.load_weights(VQA_weights_file_name)
vqa_model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
return vqa_model
K.set_image_data_format('channels_first')
from models.VQA.VQA import VQA_MODEL
from models.CNN.VGG import VGG_16
app = flask.Flask(__name__)
VQA_weights_file_name = '/input/models/VQA/VQA_MODEL_WEIGHTS.hdf5'
label_encoder_file_name = '/input/models/VQA/FULL_labelencoder_trainval.pkl'
CNN_weights_file_name = '/input/models/CNN/vgg16_weights.h5'
image_model = VGG_16(CNN_weights_file_name)
sgd = SGD(lr=0.1, decay=1e-6, momentum=0.9, nesterov=True)
image_model.compile(optimizer=sgd, loss='categorical_crossentropy')
print("Image model loaded!")
vqa_model = VQA_MODEL()
vqa_model.load_weights(VQA_weights_file_name)
vqa_model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
print("VQA Model loaded!")
def get_image_features(image_file_name):
image_features = np.zeros((1, 4096))
im = cv2.resize(cv2.imread(image_file_name), (224, 224))
#im = cv2.resize(image_file_name, (224, 224))
mean_pixel = [103.939, 116.779, 123.68]
im = im.astype(np.float32, copy=False)
for c in range(3):
im[:, :, c] = im[:, :, c] - mean_pixel[c]
im = im.transpose((2, 0, 1)) # convert the image to RGBA