def main():
options = json.loads(os.environ['SMILER_PARAMETER_MAP'])
network_string = options.get('network', 'SAM-VGG')
use_default_center_bias = options.get('center_prior',
'default') == 'default'
do_default_smoothing = options.get('do_smoothing', 'default') == 'default'
imgs_test_path = '/opt/input_vol/'
output_folder = '/opt/output_vol/'
os.makedirs(output_folder, exist_ok=True)
x = Input((3, shape_r, shape_c))
x_maps = Input((nb_gaussian, shape_r_gt, shape_c_gt))
if network_string == "SAM-VGG":
m = Model(input=[x, x_maps], output=sam_vgg([x, x_maps]))
print("Compiling SAM-VGG...")
m.compile(RMSprop(lr=1e-4),
loss=[kl_divergence, correlation_coefficient, nss])
print("Loading SAM-VGG weights...")
m.load_weights('weights/sam-vgg_salicon_weights.pkl')
elif network_string == "SAM-ResNet":
m = Model(input=[x, x_maps], output=sam_resnet([x, x_maps]))
print("Compiling SAM-ResNet...")
m.compile(RMSprop(lr=1e-4),
loss=[kl_divergence, correlation_coefficient, nss])
print("Loading SAM-ResNet weights...")
m.load_weights('weights/sam-resnet_salicon_weights.pkl')
else:
raise NotImplementedError(
"The only supported network strings are SAM-VGG and SAM-ResNet! '{}' is unknown."
.format(network_string))
gaussian = np.zeros((b_s, nb_gaussian, shape_r_gt, shape_c_gt))
def compute_saliency(image_path):
if use_default_center_bias:
predictions = m.predict(
[preprocess_images([image_path], shape_r, shape_c),
gaussian])[0]
else:
predictions = m.predict(
[preprocess_images([image_path], shape_r, shape_c),
gaussian])[0]
original_image = cv2.imread(image_path, 0)
res = postprocess_predictions(
predictions[0][0],
original_image.shape[0],
original_image.shape[1],
do_default_smoothing=do_default_smoothing)
return res
run_model(compute_saliency)
def saliencyattentivemodel(inputimage):
x = Input((3, shape_r, shape_c))
x_maps = Input((nb_gaussian, shape_r_gt, shape_c_gt))
# version (0 for SAM-VGG and 1 for SAM-ResNet)
if version == 0:
m = Model(input=[x, x_maps], output=sam_vgg([x, x_maps]))
print("Compiling SAM-VGG")
m.compile(RMSprop(lr=1e-4),
loss=[kl_divergence, correlation_coefficient, nss])
elif version == 1:
m = Model(input=[x, x_maps], output=sam_resnet([x, x_maps]))
print("Compiling SAM-ResNet")
m.compile(RMSprop(lr=1e-4),
loss=[kl_divergence, correlation_coefficient, nss])
else:
raise NotImplementedError
# Output Folder Path
output_folder = 'predictions/'
nb_imgs_test = 1 #len(file_names)
if nb_imgs_test % b_s != 0:
print(
"The number of test images should be a multiple of the batch size. Please change your batch size in config.py accordingly."
)
exit()
if version == 0:
print("Loading SAM-VGG weights")
m.load_weights('weights/sam-vgg_salicon_weights.pkl')
elif version == 1:
print("Loading SAM-ResNet weights")
m.load_weights('weights/sam-resnet_salicon_weights.pkl')
predictions = m.predict_generator(
generator_test_singleimage(b_s=b_s, image=inputimage), nb_imgs_test)[0]
print("predictions:", predictions[0])
outname = 'SalmapFromWrapper.jpg'
original_image = image
res = postprocess_predictions(predictions[0], original_image.shape[0],
original_image.shape[1])
#mport pdb; pdb.set_trace()
cv2.imwrite(output_folder + '%s' % outname, res.astype(int))
cv2.imshow('salmap', res.astype('uint8'))
cv2.waitKey(0)
return res.astype('uint8')
if __name__ == '__main__':
if len(sys.argv) == 1:
raise NotImplementedError
else:
phase = sys.argv[1]
x = Input((3, shape_r, shape_c))
x_maps = Input((nb_gaussian, shape_r_gt, shape_c_gt))
if version == 0:
m = Model(input=[x, x_maps], output=sam_vgg([x, x_maps]))
print("Compiling SAM-VGG")
m.compile(RMSprop(lr=1e-4),
loss=[kl_divergence, correlation_coefficient, nss])
elif version == 1:
m = Model(input=[x, x_maps], output=sam_resnet([x, x_maps]))
print("Compiling SAM-ResNet")
m.compile(RMSprop(lr=1e-4),
loss=[kl_divergence, correlation_coefficient, nss])
else:
raise NotImplementedError
if phase == 'train':
if nb_imgs_train % b_s != 0 or nb_imgs_val % b_s != 0:
print(
"The number of training and validation images should be a multiple of the batch size. Please change your batch size in config.py accordingly."
)
exit()
if version == 0:
print("Training SAM-VGG")
x_maps = Input((nb_gaussian, shape_r_gt, shape_c_gt))
#x_maps = Input((shape_r_gt, shape_c_gt, nb_gaussian)) #Nueva version
if version == 0: #NO USADO
# m = Model(input=[x, x_maps], output=sam_vgg([x, x_maps]))
print("Not Compiling SAM-VGG") #Nueva version
# m.compile(RMSprop(lr=1e-4), loss=[kl_divergence, correlation_coefficient, nss])
elif version == 1:
'''Hint of the problem: something is not the output of a keras layer.
You should put it in a lambda layer
When invoking the Model API, the value for outputs argument should
be tensor(or list of tensors), in this case it is a list of list of
tensors, hence there is a problem'''
#m = Model(input=[x, x_maps], output=sam_resnet([x, x_maps]))
#m = Model(inputs=[x, x_maps], outputs=sam_resnet([x, x_maps])) #New version
m = ModelAux(inputs=[x, x_maps], outputs=sam_resnet([x, x_maps])) #Final version
print("Compiling SAM-ResNet")
m.compile(RMSprop(lr=1e-4),
loss=[kl_divergence, correlation_coefficient, nss])
print("Compilado")
else:
raise NotImplementedError
if phase == 'train':
if nb_imgs_train % b_s != 0 or nb_imgs_val % b_s != 0:
print("The number of training and validation images should be a multiple of the batch size. Please change your batch size in config.py accordingly.")
exit()
if version == 0:
print("Training SAM-VGG")
cache_subdir='models')
weights_path = 'weights.sam-vgg.05-196.3380.h5'
m.load_weights(weights_path, by_name=True)
print("Compili ng SAM-VGG")
m.compile(Adam(lr=1e-4),
loss=[
kl_divergence, kl_divergence, kl_divergence,
kl_divergence, kl_divergence, kl_divergence,
kl_divergence, kl_divergence, kl_divergence,
kl_divergence
]) #
elif version == 1:
m = Model(inputs=img_input, outputs=sam_resnet(img_input))
# Load weights
weights_path = get_file(
'resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5',
RES_TF_WEIGHTS_PATH_NO_TOP,
cache_subdir='models')
print('loading weights from {}'.format(weights_path))
m.load_weights(weights_path, by_name=True)
# m.load_weights('weights.sam-resnet.06-15.6823.pkl')
# m.load_weights('weights.sam-resnet.00-17.5592.pkl', by_name=True)
print("Compiling SAM-ResNet")
m.compile(Adam(lr=1e-4),
loss=[kl_divergence, kl_divergence, kl_divergence])
# m.compile(RMSprop(lr=1e-4), loss=weighted_crossentropy)
else: