def eval_imagenet_q(net_name, param_pickle_path):
netparams = load.load_netparams_tf_q(param_pickle_path)
data_spec = networks.get_data_spec(net_name)
input_node = tf.placeholder(tf.float32,
shape=(None, data_spec.crop_size,
data_spec.crop_size,
data_spec.channels))
label_node = tf.placeholder(tf.int32)
if net_name == 'alexnet':
logits_ = networks.alexnet(input_node, netparams)
elif net_name == 'googlenet':
logits_ = networks.googlenet(input_node, netparams)
elif net_name == 'nin':
logits_ = networks.nin(input_node, netparams)
elif net_name == 'resnet18':
logits_ = networks.resnet18(input_node, netparams)
elif net_name == 'resnet50':
logits_ = networks.resnet50(input_node, netparams)
elif net_name == 'squeezenet':
logits_ = networks.squeezenet(input_node, netparams)
elif net_name == 'vgg16net':
logits_ = networks.vgg16net_noisy(input_node, netparams)
loss_op = tf.reduce_mean(
tf.nn.softmax_cross_entropy_with_logits(logits=logits_,
labels=label_node)) #
probs = softmax(logits_)
top_k_op = tf.nn.in_top_k(probs, label_node, 5)
optimizer = tf.train.AdamOptimizer(learning_rate=0.001, epsilon=0.1)
correct_pred = tf.equal(tf.argmax(probs, 1), tf.argmax(label_node, 1))
accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))
count = 0
correct = 0
cur_accuracy = 0
saver = tf.train.Saver()
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
image_producer = dataset.ImageNetProducer(val_path=IMAGE_LABLE,
data_path=IMAGE_PATH,
data_spec=data_spec)
total = len(image_producer)
coordinator = tf.train.Coordinator()
threads = image_producer.start(session=sess, coordinator=coordinator)
for (labels, images) in image_producer.batches(sess):
correct += np.sum(
sess.run(top_k_op,
feed_dict={
input_node: images,
label_node: labels
}))
count += len(labels)
cur_accuracy = float(correct) * 100 / count
#print('{:>6}/{:<6} {:>6.2f}%'.format(count, total, cur_accuracy))
print(cur_accuracy)
coordinator.request_stop()
coordinator.join(threads, stop_grace_period_secs=2)
return cur_accuracy
def get_predict_labels():
inputs = tf.placeholder("float", [None, 64, 64, 1])
is_training = tf.placeholder("bool")
prediction, _ = googlenet(inputs, is_training)
predict_labels = tf.argmax(prediction, 1)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
data = sio.loadmat("../data/dataset.mat")
testdata = data["test"] / 127.5 - 1.0
testlabel = data["testlabels"]
saver.restore(sess, "../save_para/.\\model.ckpt")
nums_test = testlabel.shape[1]
PREDICT_LABELS = np.zeros([nums_test])
for i in range(nums_test // BATCH_SIZE):
PREDICT_LABELS[i * BATCH_SIZE:i * BATCH_SIZE + BATCH_SIZE] = sess.run(predict_labels, feed_dict={inputs: testdata[i * BATCH_SIZE:i * BATCH_SIZE + BATCH_SIZE], is_training: False})
PREDICT_LABELS[(nums_test // BATCH_SIZE - 1) * BATCH_SIZE + BATCH_SIZE:] = sess.run(predict_labels, feed_dict={inputs: testdata[(nums_test // BATCH_SIZE - 1) * BATCH_SIZE + BATCH_SIZE:], is_training: False})
np.savetxt("../data/predict_labels.txt", PREDICT_LABELS)
def get_feature():
inputs = tf.placeholder("float", [None, 64, 64, 1])
is_training = tf.placeholder("bool")
_, feature = googlenet(inputs, is_training)
feature = tf.squeeze(feature, [1, 2])
sess = tf.Session()
sess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
data = sio.loadmat("../data/dataset.mat")
testdata = data["test"] / 127.5 - 1.0
testlabels = data["testlabels"]
saver.restore(sess, "../save_para/.\\model.ckpt")
nums_test = testdata.shape[0]
FEATURE = np.zeros([nums_test, 1024])
for i in range(nums_test // BATCH_SIZE):
FEATURE[i * BATCH_SIZE:i * BATCH_SIZE + BATCH_SIZE] = sess.run(feature, feed_dict={inputs: testdata[i * BATCH_SIZE:i * BATCH_SIZE + BATCH_SIZE], is_training: False})
FEATURE[(nums_test // BATCH_SIZE - 1) * BATCH_SIZE + BATCH_SIZE:] = sess.run(feature, feed_dict={inputs: testdata[(nums_test // BATCH_SIZE - 1) * BATCH_SIZE + BATCH_SIZE:], is_training: False})
sio.savemat("../data/feature.mat", {"feature": FEATURE, "testlabels": testlabels})
