def test(images_tensor, labels_tensor, parameters_path, test_data_dir):
'''
测试模型的有效性
:param images_tensor: 图像的tensor
:param labels_tensor: label的tensor
:param generator: 数据的生成器，可以通过for (images_batch, labels_batch) in generator:格式来获取数据
:param parameters_path:模型保存的路径
:return:
'''
logits = inference(images_tensor, False)
prediction_tensor = tf.argmax(logits, 1)
correct_prediction = tf.equal(tf.argmax(logits, 1),
tf.cast(tf.squeeze(labels_tensor), tf.int64))
accuracy_tensor = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
saver = tf.train.Saver()
featuremaps = []
labels = []
with tf.Session() as sess:
full_path = tf.train.latest_checkpoint(parameters_path)
print full_path
saver.restore(sess, full_path)
names = os.listdir(test_data_dir)
paths = []
labels = []
all_predictions = []
for name in names:
if int(name[-1]) not in [0, 1, 2, 3]:
continue
paths.append(os.path.join(test_data_dir, name))
labels.append(int(name[-1]))
start_index = 0
while True:
end_index = start_index + net_config.BATCH_SIZE
if end_index > len(paths):
end_index = len(paths)
cur_batch_paths = paths[start_index:end_index]
cur_batch_images = [
load_ROI(cur_path) for cur_path in cur_batch_paths
]
cur_batch_labels = labels[start_index:end_index]
cur_batch_images = resize_images(cur_batch_images,
net_config.ROI_SIZE_W, True)
print np.shape(cur_batch_images)
print np.shape(cur_batch_labels)
predicted, accuracy = sess.run(
[prediction_tensor, accuracy_tensor],
feed_dict={
images_tensor: np.array(cur_batch_images),
labels_tensor: np.squeeze(cur_batch_labels)
})
print accuracy
all_predictions.extend(predicted)
start_index = end_index
if start_index >= len(paths):
break
calculate_acc_error(all_predictions, labels)
def execute_classify(train_features, train_labels, val_features, val_labels, test_features, test_labels):
from LeaningBased.BoVW_DualDict.classification import SVM, LinearSVM, KNN
predicted_label, c_params, g_params, max_c, max_g, accs = SVM.do(train_features, train_labels, val_features,
val_labels,
adjust_parameters=True)
predicted_label, accs = SVM.do(train_features, train_labels, test_features, test_labels,
adjust_parameters=False, C=max_c, gamma=max_g)
calculate_acc_error(predicted_label, test_labels)
print 'ACA is ', accs
return accs
from utils.classification import SVM, KNN
train_data = scio.loadmat(
'./features/crossvalidation/0/saved/train.npy.mat')
train_features = train_data['features']
train_labels = train_data['labels']
val_data = scio.loadmat('./features/crossvalidation/0/saved/val.npy.mat')
val_features = val_data['features']
val_labels = val_data['labels']
test_data = scio.loadmat('./features/crossvalidation/0/saved/test.npy.mat')
test_features = test_data['features']
test_labels = test_data['labels']
# SVM
predicted_label, c_params, g_params, accs = SVM.do(train_features,
train_labels,
val_features,
val_labels,
adjust_parameters=True)
# use default parameters
predicted_label, acc = SVM.do(train_features,
train_labels,
test_features,
test_labels,
adjust_parameters=False,
C=c_params,
gamma=g_params)
print 'ACC is ', acc
calculate_acc_error(predicted_label, test_labels)
def main(_):
roi_images = tf.placeholder(shape=[
None, net_config.ROI_SIZE_W, net_config.ROI_SIZE_H,
net_config.IMAGE_CHANNEL
],
dtype=np.float32,
name='roi_input')
expand_roi_images = tf.placeholder(shape=[
None, net_config.EXPAND_SIZE_W, net_config.EXPAND_SIZE_H,
net_config.IMAGE_CHANNEL
],
dtype=np.float32,
name='expand_roi_input')
batch_size_tensor = tf.placeholder(dtype=tf.int32, shape=[])
is_training_tensor = tf.placeholder(dtype=tf.bool, shape=[])
logits, _, _, _ = inference_small(roi_images,
expand_roi_images,
phase_names=['NC', 'ART', 'PV'],
num_classes=4,
is_training=is_training_tensor,
batch_size=batch_size_tensor)
model_path = '/home/give/PycharmProjects/MICCAI2018/deeplearning/Parallel/parameters/1'
# model_path = '/home/give/PycharmProjects/MedicalImage/Net/forpatch/cross_validation/model/multiscale/parallel/0/2200.0'
predictions = tf.nn.softmax(logits)
saver = tf.train.Saver(tf.all_variables())
print predictions
predicted_label_tensor = tf.argmax(predictions, axis=1)
print predicted_label_tensor
init = tf.initialize_all_variables()
sess = tf.Session(config=tf.ConfigProto(log_device_placement=False))
sess.run(init)
tf.train.start_queue_runners(sess=sess)
latest = tf.train.latest_checkpoint(model_path)
if not latest:
print "No checkpoint to continue from in", model_path
sys.exit(1)
print "resume", latest
saver.restore(sess, latest)
data_dir = '/home/give/Documents/dataset/MICCAI2018/Patches/crossvalidation/1/test'
slice_dir = '/home/give/Documents/dataset/MICCAI2018/Slices/crossvalidation/1/test'
labels = []
paths = []
for typeid in [0, 1, 2, 3]:
cur_path = os.path.join(data_dir, str(typeid))
names = os.listdir(cur_path)
labels.extend([typeid] * len(names))
paths.extend([os.path.join(cur_path, name) for name in names])
paths, labels = shuffle_image_label(paths, labels)
start_index = 0
predicted_labels = []
liver_density = load_raw_liver_density()
while True:
if start_index >= len(paths):
break
print start_index, len(paths)
end_index = start_index + net_config.BATCH_SIZE
cur_paths = paths[start_index:end_index]
cur_roi_images = [np.asarray(load_patch(path)) for path in cur_paths]
cur_expand_roi_images = [
np.asarray(load_patch(path, return_roi=True, parent_dir=slice_dir))
for path in cur_paths
]
cur_roi_images = resize_images(cur_roi_images, net_config.ROI_SIZE_W,
True)
cur_expand_roi_images = resize_images(cur_expand_roi_images,
net_config.EXPAND_SIZE_W, True)
cur_liver_densitys = [
liver_density[os.path.basename(path)
[:os.path.basename(path).rfind('_')]]
for path in cur_paths
]
# for i in range(len(cur_roi_images)):
# for j in range(3):
# cur_roi_images[i, :, :, j] = (1.0 * cur_roi_images[i, :, :, j]) / (1.0 * cur_liver_densitys[i][j])
# cur_expand_roi_images[i, :, :, j] = (1.0 * cur_expand_roi_images[i, :, :, j]) / (
# 1.0 * cur_liver_densitys[i][j])
predicted_batch_labels = sess.run(predicted_label_tensor,
feed_dict={
roi_images:
cur_roi_images,
expand_roi_images:
cur_expand_roi_images,
is_training_tensor:
False,
batch_size_tensor:
len(cur_roi_images)
})
batch_labels = labels[start_index:end_index]
predicted_labels.extend(predicted_batch_labels)
start_index = end_index
calculate_acc_error(predicted_batch_labels, batch_labels)
calculate_acc_error(predicted_labels, labels)
@staticmethod
def do(train_data, train_label, test_data, test_label=None, adjust_parameters=True):
train_data = np.array(train_data).squeeze()
train_label = np.array(train_label).squeeze()
test_data = np.array(test_data).squeeze()
if test_label is not None:
test_label = np.array(test_label).squeeze()
svm = LinearSVC()
svm.fit(train_data, train_label)
predicts = svm.predict(test_data)
acc = None
if test_label is not None:
acc = accuracy_score(test_label, predicts)
print acc
return predicts
if __name__ == '__main__':
data = scio.loadmat('/home/give/PycharmProjects/MedicalImage/BoVW/data_256_False.mat')
train_features = data['train_features']
val_features = data['val_features']
train_labels = data['train_labels']
val_labels = data['val_labels']
val_labels = np.squeeze(val_labels)
print np.shape(train_features), np.shape(train_labels)
predicted_label = SVM.do(train_features, train_labels, val_features, val_labels, adjust_parameters=True)
np.save('./predicted_res.npy', predicted_label)
# predicted_label = np.load('./predicted_res.npy')
calculate_acc_error(predicted_label, val_labels)
def train(logits,
local_output_tensor,
global_output_tensor,
represent_feature_tensor,
images_tensor,
expand_images_tensor,
labels_tensor,
is_training_tensor,
save_model_path=None,
step_width=100,
record_loss=False):
cross_id = 1
has_centerloss = False
patches_dir = '/home/give/Documents/dataset/MICCAI2018/Patches/crossvalidation'
roi_dir = '/home/give/Documents/dataset/MICCAI2018/Slices/crossvalidation'
pre_load = True
train_dataset = DataSet(os.path.join(patches_dir, str(cross_id), 'train'),
'train',
pre_load=pre_load,
rescale=True,
divied_liver=False,
expand_is_roi=True,
full_roi_path=os.path.join(roi_dir, str(cross_id),
'train'))
val_dataset = DataSet(os.path.join(patches_dir, str(cross_id), 'val'),
'val',
pre_load=pre_load,
rescale=True,
divied_liver=False,
expand_is_roi=True,
full_roi_path=os.path.join(roi_dir, str(cross_id),
'val'))
train_batchdata = train_dataset.get_next_batch(net_config.BATCH_SIZE)
val_batchdata = val_dataset.get_next_batch(net_config.BATCH_SIZE)
global_step = tf.get_variable('global_step', [],
initializer=tf.constant_initializer(0),
trainable=False)
val_step = tf.get_variable('val_step', [],
initializer=tf.constant_initializer(0),
trainable=False)
# inter loss
loss_local = loss(local_output_tensor, labels_tensor)
loss_global = loss(global_output_tensor, labels_tensor)
loss_last = loss(logits, labels_tensor)
loss_inter = loss_local_coefficient * loss_local + loss_global_coefficient * loss_global * loss_all_coefficient * loss_last
# intra loss
if has_centerloss:
represent_feature_tensor_shape = represent_feature_tensor.get_shape(
).as_list()
print 'represent_feature_tensor_shape is ', represent_feature_tensor_shape
centers_value = np.zeros(
[category_num, represent_feature_tensor_shape[1]],
dtype=np.float32)
print 'centers_value shape is ', np.shape(centers_value)
centers_tensor = tf.placeholder(
dtype=tf.float32,
shape=[category_num, represent_feature_tensor_shape[1]])
print 'center_tensor shape is ', tf.shape(centers_tensor)
center_loss = calculate_centerloss(represent_feature_tensor,
labels_tensor,
centers_tensor=centers_tensor)
owner_step = tf.py_func(update_centers, [
centers_tensor, represent_feature_tensor, labels_tensor,
category_num
], tf.float32)
loss_ = loss_inter + _lambda * center_loss
else:
loss_ = loss_inter
predictions = tf.nn.softmax(logits)
print 'predictions shape is ', predictions
print 'label is ', labels_tensor
top1_error = top_k_error(predictions, labels_tensor, 1)
labels_onehot = tf.one_hot(labels_tensor, logits.get_shape().as_list()[-1])
print 'output node is ', logits.get_shape().as_list()[-1]
accuracy_tensor = calculate_accuracy(predictions, labels_onehot)
# loss_avg
ema = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY, global_step)
tf.add_to_collection(UPDATE_OPS_COLLECTION, ema.apply([loss_]))
tf.summary.scalar('loss_avg', ema.average(loss_))
# validation stats
ema = tf.train.ExponentialMovingAverage(0.9, val_step)
val_op = tf.group(val_step.assign_add(1), ema.apply([top1_error]))
top1_error_avg = ema.average(top1_error)
tf.summary.scalar('val_top1_error_avg', top1_error_avg)
tf.summary.scalar('learning_rate', FLAGS.learning_rate)
opt = tf.train.MomentumOptimizer(FLAGS.learning_rate, MOMENTUM)
grads = opt.compute_gradients(loss_)
for grad, var in grads:
if grad is not None and not FLAGS.minimal_summaries:
tf.summary.histogram(var.op.name + '/gradients', grad)
apply_gradient_op = opt.apply_gradients(grads, global_step=global_step)
if not FLAGS.minimal_summaries:
# Display the training images in the visualizer.
tf.summary.image('images', images_tensor)
for var in tf.trainable_variables():
tf.summary.image(var.op.name, var)
batchnorm_updates = tf.get_collection(UPDATE_OPS_COLLECTION)
batchnorm_updates_op = tf.group(*batchnorm_updates)
if has_centerloss:
with tf.control_dependencies(
[apply_gradient_op, batchnorm_updates_op, owner_step]):
train_op = tf.no_op('train')
else:
train_op = tf.group(apply_gradient_op, batchnorm_updates_op)
saver = tf.train.Saver(tf.all_variables())
summary_op = tf.summary.merge_all()
init = tf.initialize_all_variables()
sess = tf.Session(config=tf.ConfigProto(log_device_placement=False))
sess.run(init)
tf.train.start_queue_runners(sess=sess)
summary_writer = tf.summary.FileWriter(FLAGS.log_dir, sess.graph)
val_summary_writer = tf.summary.FileWriter(FLAGS.log_val_dir, sess.graph)
if FLAGS.resume:
latest = tf.train.latest_checkpoint(FLAGS.load_model_path)
if not latest:
print "No checkpoint to continue from in", FLAGS.train_dir
sys.exit(1)
print "resume", latest
saver.restore(sess, latest)
for x in xrange(FLAGS.max_steps + 1):
start_time = time.time()
step = sess.run(global_step)
if has_centerloss:
i = [train_op, loss_, owner_step]
else:
i = [train_op, loss_]
write_summary = step % 100 and step > 1
if write_summary:
i.append(summary_op)
train_roi_batch_images, train_expand_roi_batch_images, train_labels = train_batchdata.next(
)
o = sess.run(
i,
feed_dict={
images_tensor: train_roi_batch_images,
expand_images_tensor: train_expand_roi_batch_images,
labels_tensor: train_labels,
# centers_tensor: centers_value,
is_training_tensor: True
})
if has_centerloss:
centers_value = o[2]
loss_value = o[1]
duration = time.time() - start_time
assert not np.isnan(loss_value), 'Model diverged with loss = NaN'
if (step - 1) % step_width == 0:
top1_error_value, accuracy_value, labels_values, predictions_values = sess.run(
[top1_error, accuracy_tensor, labels_tensor, predictions],
feed_dict={
images_tensor: train_roi_batch_images,
expand_images_tensor: train_expand_roi_batch_images,
labels_tensor: train_labels,
# centers_tensor: centers_value,
is_training_tensor: True
})
predictions_values = np.argmax(predictions_values, axis=1)
examples_per_sec = FLAGS.batch_size / float(duration)
# accuracy = eval_accuracy(predictions_values, labels_values)
format_str = (
'step %d, loss = %.2f, top1 error = %g, accuracy value = %g (%.1f examples/sec; %.3f '
'sec/batch)')
print(format_str % (step, loss_value, top1_error_value,
accuracy_value, examples_per_sec, duration))
if write_summary:
if has_centerloss:
summary_str = o[3]
else:
summary_str = o[2]
summary_writer.add_summary(summary_str, step)
# Save the model checkpoint periodically.
if step > 1 and step % step_width == 0:
checkpoint_path = os.path.join(save_model_path, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=global_step)
save_dir = os.path.join(save_model_path, str(step))
if not os.path.exists(save_dir):
os.mkdir(save_dir)
filenames = glob(
os.path.join(save_model_path,
'*-' + str(int(step + 1)) + '.*'))
for filename in filenames:
shutil.copy(filename,
os.path.join(save_dir, os.path.basename(filename)))
# Run validation periodically
if step > 1 and step % step_width == 0:
val_roi_batch_images, val_expand_roi_batch_images, val_labels = val_batchdata.next(
)
_, top1_error_value, summary_value, accuracy_value, labels_values, predictions_values = sess.run(
[
val_op, top1_error, summary_op, accuracy_tensor,
labels_tensor, predictions
],
{
images_tensor: val_roi_batch_images,
expand_images_tensor: val_expand_roi_batch_images,
# centers_tensor: centers_value,
labels_tensor: val_labels,
is_training_tensor: False
})
predictions_values = np.argmax(predictions_values, axis=1)
# accuracy = eval_accuracy(predictions_values, labels_values)
calculate_acc_error(logits=predictions_values,
label=labels_values,
show=True)
print('Validation top1 error %.2f, accuracy value %f' %
(top1_error_value, accuracy_value))
val_summary_writer.add_summary(summary_value, step)
def main(_):
roi_images = tf.placeholder(shape=[
None, net_config.ROI_SIZE_W, net_config.ROI_SIZE_H,
net_config.IMAGE_CHANNEL
],
dtype=np.float32,
name='roi_input')
expand_roi_images = tf.placeholder(shape=[
None, net_config.EXPAND_SIZE_W, net_config.EXPAND_SIZE_H,
net_config.IMAGE_CHANNEL
],
dtype=np.float32,
name='expand_roi_input')
batch_size_tensor = tf.placeholder(dtype=tf.int32, shape=[])
is_training_tensor = tf.placeholder(dtype=tf.bool, shape=[])
logits = inference_small(roi_images,
expand_roi_images,
phase_names=['NC', 'ART', 'PV'],
num_classes=4,
point_phase=[2],
is_training=is_training_tensor,
batch_size=batch_size_tensor)
# model_path = '/home/give/PycharmProjects/MedicalImage/Net/ICIP/4-class/Patch_ROI/models/300.0/'
# model_path = '/home/give/PycharmProjects/MedicalImage/Net/forpatch/cross_validation/model/multiscale/parallel/0/2200.0'
model_path = '/home/give/PycharmProjects/MedicalImage/Net/ICIP/4-class/Patch_ROI/models_7'
predictions = tf.nn.softmax(logits)
saver = tf.train.Saver(tf.all_variables())
print predictions
predicted_label_tensor = tf.argmax(predictions, axis=1)
print predicted_label_tensor
init = tf.initialize_all_variables()
sess = tf.Session(config=tf.ConfigProto(log_device_placement=False))
sess.run(init)
tf.train.start_queue_runners(sess=sess)
latest = tf.train.latest_checkpoint(model_path)
if not latest:
print "No checkpoint to continue from in", model_path
sys.exit(1)
print "resume", latest
saver.restore(sess, latest)
data_dir = '/home/give/Documents/dataset/MedicalImage/MedicalImage/Patches/ICIP/only-patch-7/val'
labels = []
paths = []
mapping_label = {0: 0, 1: 1, 2: 2, 3: 3}
for typeid in [0, 1, 2, 3]:
cur_path = os.path.join(data_dir, str(typeid))
names = os.listdir(cur_path)
labels.extend([mapping_label[typeid]] * len(names))
paths.extend([os.path.join(cur_path, name) for name in names])
paths, labels = shuffle_image_label(paths, labels)
start_index = 0
predicted_labels = []
while True:
if start_index >= len(paths):
break
print start_index, len(paths)
end_index = start_index + net_config.BATCH_SIZE
cur_paths = paths[start_index:end_index]
cur_roi_images = [np.asarray(load_patch(path)) for path in cur_paths]
cur_expand_roi_images = [
np.asarray(
load_patch(
path,
return_roi=True,
parent_dir=
'/home/give/Documents/dataset/MedicalImage/MedicalImage/SL_TrainAndVal/val'
)) for path in cur_paths
]
cur_roi_images = resize_images(cur_roi_images, net_config.ROI_SIZE_W,
True)
cur_expand_roi_images = resize_images(cur_expand_roi_images,
net_config.EXPAND_SIZE_W, True)
predicted_batch_labels = sess.run(predicted_label_tensor,
feed_dict={
roi_images:
cur_roi_images,
expand_roi_images:
cur_expand_roi_images,
is_training_tensor:
False,
batch_size_tensor:
len(cur_roi_images)
})
batch_labels = labels[start_index:end_index]
predicted_labels.extend(predicted_batch_labels)
start_index = end_index
calculate_acc_error(predicted_batch_labels, batch_labels)
calculate_acc_error(predicted_labels, labels)
def train(logits,
images_tensor,
expand_images_tensor,
labels_tensor,
is_training_tensor,
save_model_path=None,
step_width=100,
record_loss=False):
train_dataset = DataSet(
'/home/give/Documents/dataset/MedicalImage/MedicalImage/Patches/ICIP/140_only_patch/train',
'train',
rescale=True,
divied_liver=False,
expand_is_roi=True,
full_roi_path=
'/home/give/Documents/dataset/MedicalImage/MedicalImage/SL_TrainAndVal/train'
)
val_dataset = DataSet(
'/home/give/Documents/dataset/MedicalImage/MedicalImage/Patches/ICIP/140_only_patch/val',
'val',
rescale=True,
divied_liver=False,
expand_is_roi=True,
full_roi_path=
'/home/give/Documents/dataset/MedicalImage/MedicalImage/SL_TrainAndVal/train'
)
loss_value_record_file_path = '/home/give/PycharmProjects/MedicalImage/Net/ICIP/4-class/Patch_ROI/models_draw_line/loss_value'
if record_loss:
writed_filed = open(loss_value_record_file_path, 'w')
writed_filed.write(
'step training loss, val loss, training acc, val acc\n')
writed_filed.close()
train_batchdata = train_dataset.get_next_batch(net_config.BATCH_SIZE)
val_batchdata = val_dataset.get_next_batch(net_config.BATCH_SIZE)
global_step = tf.get_variable('global_step', [],
initializer=tf.constant_initializer(0),
trainable=False)
val_step = tf.get_variable('val_step', [],
initializer=tf.constant_initializer(0),
trainable=False)
loss_ = loss(logits, labels_tensor)
predictions = tf.nn.softmax(logits)
print 'predictions shape is ', predictions
print 'label is ', labels_tensor
top1_error = top_k_error(predictions, labels_tensor, 1)
labels_onehot = tf.one_hot(labels_tensor, logits.get_shape().as_list()[-1])
print 'output node is ', logits.get_shape().as_list()[-1]
accuracy_tensor = calculate_accuracy(predictions, labels_onehot)
# loss_avg
ema = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY, global_step)
tf.add_to_collection(UPDATE_OPS_COLLECTION, ema.apply([loss_]))
tf.summary.scalar('loss_avg', ema.average(loss_))
# validation stats
ema = tf.train.ExponentialMovingAverage(0.9, val_step)
val_op = tf.group(val_step.assign_add(1), ema.apply([top1_error]))
top1_error_avg = ema.average(top1_error)
tf.summary.scalar('val_top1_error_avg', top1_error_avg)
tf.summary.scalar('learning_rate', FLAGS.learning_rate)
opt = tf.train.MomentumOptimizer(FLAGS.learning_rate, MOMENTUM)
grads = opt.compute_gradients(loss_)
for grad, var in grads:
if grad is not None and not FLAGS.minimal_summaries:
tf.summary.histogram(var.op.name + '/gradients', grad)
apply_gradient_op = opt.apply_gradients(grads, global_step=global_step)
if not FLAGS.minimal_summaries:
# Display the training images in the visualizer.
tf.summary.image('images', images_tensor)
for var in tf.trainable_variables():
tf.summary.image(var.op.name, var)
batchnorm_updates = tf.get_collection(UPDATE_OPS_COLLECTION)
batchnorm_updates_op = tf.group(*batchnorm_updates)
train_op = tf.group(apply_gradient_op, batchnorm_updates_op)
saver = tf.train.Saver(tf.all_variables())
summary_op = tf.summary.merge_all()
init = tf.initialize_all_variables()
sess = tf.Session(config=tf.ConfigProto(log_device_placement=False))
sess.run(init)
tf.train.start_queue_runners(sess=sess)
summary_writer = tf.summary.FileWriter(FLAGS.log_dir, sess.graph)
val_summary_writer = tf.summary.FileWriter(FLAGS.log_val_dir, sess.graph)
if FLAGS.resume:
latest = tf.train.latest_checkpoint(FLAGS.load_model_path)
if not latest:
import sys
print "No checkpoint to continue from in", FLAGS.train_dir
sys.exit(1)
print "resume", latest
saver.restore(sess, latest)
for x in xrange(FLAGS.max_steps + 1):
start_time = time.time()
step = sess.run(global_step)
i = [train_op, loss_, accuracy_tensor]
write_summary = step % 100 and step > 1
if write_summary:
i.append(summary_op)
train_roi_batch_images, train_expand_roi_batch_images, train_labels = train_batchdata.next(
)
o = sess.run(i,
feed_dict={
images_tensor: train_roi_batch_images,
expand_images_tensor: train_expand_roi_batch_images,
labels_tensor: train_labels,
is_training_tensor: True
})
loss_value = o[1]
# calculate the value of loss of validation
if record_loss:
val_roi_batch_images, val_expand_roi_batch_images, val_labels = val_batchdata.next(
)
val_loss_value, val_accuracy_value = sess.run(
[loss_, accuracy_tensor],
feed_dict={
images_tensor: train_roi_batch_images,
expand_images_tensor: train_expand_roi_batch_images,
labels_tensor: train_labels,
is_training_tensor: False
})
printed_str = 'step %d, training loss_value %f accuracy %f, validation loss_value %f accuracy %f'\
% (step, loss_value, o[2], val_loss_value, val_accuracy_value)
print printed_str
writed_str = '%d %f %f %f %f\n' % (
step, loss_value, val_loss_value, o[2], val_accuracy_value)
writed_filed = open(loss_value_record_file_path, 'a+')
writed_filed.write(writed_str)
writed_filed.close()
duration = time.time() - start_time
assert not np.isnan(loss_value), 'Model diverged with loss = NaN'
if (step - 1) % step_width == 0:
top1_error_value, accuracy_value, labels_values, predictions_values = sess.run(
[top1_error, accuracy_tensor, labels_tensor, predictions],
feed_dict={
images_tensor: train_roi_batch_images,
expand_images_tensor: train_expand_roi_batch_images,
labels_tensor: train_labels,
is_training_tensor: True
})
predictions_values = np.argmax(predictions_values, axis=1)
examples_per_sec = FLAGS.batch_size / float(duration)
# accuracy = eval_accuracy(predictions_values, labels_values)
format_str = (
'step %d, loss = %.2f, top1 error = %g, accuracy value = %g (%.1f examples/sec; %.3f '
'sec/batch)')
print(format_str % (step, loss_value, top1_error_value,
accuracy_value, examples_per_sec, duration))
if write_summary:
summary_str = o[3]
summary_writer.add_summary(summary_str, step)
if step > 1 and step % step_width == 0:
checkpoint_path = os.path.join(save_model_path, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=global_step)
save_dir = os.path.join(save_model_path, str(step))
if not os.path.exists(save_dir):
os.mkdir(save_dir)
filenames = glob(
os.path.join(save_model_path,
'*-' + str(int(step + 1)) + '.*'))
for filename in filenames:
shutil.copy(filename,
os.path.join(save_dir, os.path.basename(filename)))
# Run validation periodically
if step > 1 and step % step_width == 0:
if not record_loss:
val_roi_batch_images, val_expand_roi_batch_images, val_labels = val_batchdata.next(
)
_, top1_error_value, summary_value, accuracy_value, labels_values, predictions_values = sess.run(
[
val_op, top1_error, summary_op, accuracy_tensor,
labels_tensor, predictions
], {
images_tensor: val_roi_batch_images,
expand_images_tensor: val_expand_roi_batch_images,
labels_tensor: val_labels,
is_training_tensor: False
})
predictions_values = np.argmax(predictions_values, axis=1)
# accuracy = eval_accuracy(predictions_values, labels_values)
calculate_acc_error(logits=predictions_values,
label=labels_values,
show=True)
print('Validation top1 error %.2f, accuracy value %f' %
(top1_error_value, accuracy_value))
val_summary_writer.add_summary(summary_value, step)
