labels.append(labelStore[i])
path.append(str(i) + ".JPEG")
idx += 1
if idx == batch_size:
yield labels, images, path
labels = []
path = []
images = np.zeros(batch_shape)
idx = 0
if idx > 0:
yield labels, images, path
# Load the Inception model for attributions.
attr_sess, attr_graph = load_model(MODEL_LOC)
inception_predictions_and_gradients = make_predictions_and_gradients(
attr_sess, attr_graph)
attr_labels = load_labels_vocabulary(LABELS_LOC)
image_iterator = load_images(imagePathOrig, batch_shape)
attr_iterator = load_images(attrPath, batch_shape)
power = [75]
print(len(power))
counterSame = 0
counterPert = 0
crossEntropyStore = np.zeros((len(power), 1000))
countStore = np.zeros((len(power), 1000))
sameFlag = np.zeros(1000)
flipProbStore = np.zeros((len(power), 1000))
from InceptionModel.inception_utils import load_model, load_labels_vocabulary, make_predictions_and_gradients, top_label_id_and_score
from IntegratedGradients.integrated_gradients import integrated_gradients, random_baseline_integrated_gradients
from VisualizationLibrary.visualization_lib import Visualize, show_pil_image, pil_image
MODEL_LOC='./InceptionModel/tensorflow_inception_graph.pb'
LABELS_LOC='./InceptionModel/imagenet_comp_graph_label_strings.txt'
# Load the Inception model.
sess, graph = load_model(MODEL_LOC)
# Load the Labels vocabulary.
labels = load_labels_vocabulary(LABELS_LOC)
# Make the predictions_and_gradients function
inception_predictions_and_gradients = make_predictions_and_gradients(sess, graph)
def load_image(img_path):
# "/Users/pin-jutien/Integrated-Gradients/Images/70bfca4555cca92e.jpg"
image = tf.image.decode_jpeg(tf.read_file(img_path), channels=3)
img = sess.run(image)
return img
img = load_image("/Users/pin-jutien/Integrated-Gradients/Images/70bfca4555cca92e.jpg")
# Determine top label and score.
top_label_id, score = top_label_id_and_score(img, inception_predictions_and_gradients)
print("Top label: %s, score: %f" % (labels[top_label_id], score))