def load_batch(dataset, batch_size=32, height=28, width=28, is_training=False):
data_provider = slim.dataset_data_provider.DatasetDataProvider(dataset)
image, label = data_provider.get(['image', 'label'])
image = lenet_preprocessing.preprocess_image(image, height, width,
is_training)
images, labels = tf.train.batch([image, label],
batch_size=batch_size,
allow_smaller_final_batch=True)
return images, labels
def load_batch(dataset, batch_size=32, height=28, width=28, is_training=False):
data_provider = slim.dataset_data_provider.DatasetDataProvider(dataset)
image, label = data_provider.get(['image', 'label'])
image = lenet_preprocessing.preprocess_image(
image,
height,
width,
is_training)
images, labels = tf.train.batch(
[image, label],
batch_size=batch_size,
allow_smaller_final_batch=True)
return images, labels
def classify_image(filepath):
with tf.Graph().as_default():
image = open(filepath, 'rb')
# Open specified url and load image as a string
image_string = image.read()
# Decode string into matrix with intensity values
image = tf.image.decode_png(image_string, channels=3)
# Resize the input image, preserving the aspect ratio
# and make a central crop of the resulted image.
# The crop will be of the size of the default image size of
# the network.
processed_image = lenet_preprocessing.preprocess_image(
image, image_size, image_size, is_training=False)
# Networks accept images in batches.
# The first dimension usually represents the batch size.
# In our case the batch size is one.
processed_images = tf.expand_dims(processed_image, 0)
# Create the model, use the default arg scope to configure
# the batch norm parameters. arg_scope is a very convenient
# feature of slim library -- you can define default
# parameters for layers -- like stride, padding etc.
with slim.arg_scope(lenet.lenet_arg_scope()):
logits, _ = lenet.lenet(processed_images,
num_classes=11,
is_training=False)
# In order to get probabilities we apply softmax on the output.
probabilities = tf.nn.softmax(logits)
# Create a function that reads the network weights
# from the checkpoint file that you downloaded.
# We will run it in session later.
init_fn = slim.assign_from_checkpoint_fn(
os.path.join(checkpoints_dir, 'model.ckpt-10000'),
slim.get_model_variables(None))
with tf.Session() as sess:
# Load weights
init_fn(sess)
# We want to get predictions, image as numpy matrix
# and resized and cropped piece that is actually
# being fed to the network.
np_image, network_input, probabilities = sess.run(
[image, processed_image, probabilities])
probabilities = probabilities[0, 0:]
sorted_inds = [
i[0]
for i in sorted(enumerate(-probabilities), key=lambda x: x[1])
]
for i in range(11):
index = sorted_inds[i]
print('Probability %0.2f => [%s]' %
(probabilities[index], names[index]))
return sorted_inds[0], probabilities