def predict(self):
self.predict_setup()
self.sess.run(tf.global_variables_initializer())
self.sess.run(tf.local_variables_initializer())
# load checkpoint
checkpointfile = self.conf.modeldir + '/model.ckpt-' + str(self.conf.valid_step)
self.load(self.loader, checkpointfile)
# Start queue threads.
threads = tf.train.start_queue_runners(coord=self.coord, sess=self.sess)
# img_name_list
image_list, _ = read_labeled_image_list('', self.conf.test_data_list)
# Predict!
for step in range(self.conf.test_num_steps):
preds = self.sess.run(self.pred)
predmatrix = preds[0, :, :, 0, :]
check = np.sum(predmatrix,axis=2)
for i in range(self.conf.num_classes):
img_name = image_list[step].split('/')[2].split('.')[0]
# Save raw predictions, i.e. each pixel is an integer between [0,20].
im = Image.fromarray(preds[0, :, :, 0, i], mode='L')
if not os.path.isdir(self.conf.out_dir + "/prediction/" + str(i)):
os.mkdir(self.conf.out_dir + "/prediction/" + str(i))
filename = '/%s_mask.png' % (img_name)
im.save(self.conf.out_dir + '/prediction/' + str(i) + filename)
# Save predictions for visualization.
# See utils/label_utils.py for color setting
# Need to be modified based on datasets.
if self.conf.visual:
#msk = decode_labels(preds[:, :, :, :, i], num_classes=self.conf.num_classes) #old version
#im = Image.fromarray(msk[0], mode='RGB') #old version
if not os.path.isdir(self.conf.out_dir + '/visual_prediction/' + str(i)):
os.mkdir(self.conf.out_dir + '/visual_prediction/' + str(i))
filename = '/%s_mask_visual.png' % (img_name)
print("saving Image")
#im.save(self.conf.out_dir + '/visual_prediction/' + str(i) + filename) #old version
plt.imsave(self.conf.out_dir + '/visual_prediction/' + str(i) + filename, preds[0, :, :, 0, i])
if step % 100 == 0:
print('step {:d}'.format(step))
print('The output files has been saved to {}'.format(self.conf.out_dir))
# finish
self.coord.request_stop()
self.coord.join(threads)
def predict(self):
normal_color = "\033[0;37;40m"
self.predict_setup()
self.sess.run(tf.global_variables_initializer())
self.sess.run(tf.local_variables_initializer())
# load checkpoint
checkpointfile = self.conf.modeldir + '/model.ckpt-' + str(
self.conf.test_step)
self.load(self.loader, checkpointfile)
# Start queue threads.
threads = tf.train.start_queue_runners(coord=self.coord,
sess=self.sess)
# img_name_list
image_list, _ = read_labeled_image_list('', self.conf.test_data_list)
# Predict!
for step in range(self.conf.test_num_steps):
preds, raw_preds = self.sess.run([self.pred, raw_output_])
img_name = image_list[step].split('/')[2].split('.')[0]
# Save raw predictions, i.e. each pixel is an integer between [0,20].
prior1 = self.conf.prior
im = Image.fromarray(preds[0, :, :, 0], mode='L')
filename = '/%s_mask.png' % (img_name)
im.save(self.conf.out_dir + '/prediction' + '/' + str(prior1) +
filename)
# Save predictions for visualization.
# See utils/label_utils.py for color setting
# Need to be modified based on datasets.
if self.conf.visual:
msk = decode_labels(preds, num_classes=self.conf.num_classes)
im = Image.fromarray(msk[0], mode='RGB')
filename = '/%s_mask_visual.png' % (img_name)
im.save(self.conf.out_dir + '/visual_prediction' + '/' +
str(prior1) + filename)
if step % 100 == 0:
print('step {:d}'.format(step))
print(
'The output files has been saved to {}'.format(self.conf.out_dir) +
normal_color)
# finish
self.coord.request_stop()
self.coord.join(threads)
def predict(self):
self.predict_setup()
self.sess.run(tf.global_variables_initializer())
self.sess.run(tf.local_variables_initializer())
# load checkpoint
#checkpointfile = self.conf.modeldir+ '/model.ckpt-' + str(self.conf.valid_step)
#checkpointfile = 'deeplab_resnet_init.ckpt'
checkpointfile = tf.train.latest_checkpoint("./model_multigpu_bs10/")
#checkpointfile = './model_crf_test0'+ '/model.ckpt-' + '0'
self.load(self.loader, checkpointfile)
# Start queue threads.
threads = tf.train.start_queue_runners(coord=self.coord,
sess=self.sess)
# img_name_list
image_list, _ = read_labeled_image_list('', self.conf.test_data_list)
# Predict!
for step in range(self.conf.test_num_steps):
preds = self.sess.run(self.pred)
resized_dec = self.sess.run(self.resized_decoder100)
img_name = image_list[step].split('/')[2].split('.')[0]
# Save raw predictions, i.e. each pixel is an integer between [0,20].
im = Image.fromarray(preds[0, :, :, 0], mode='L')
filename = '/%s_mask.png' % (img_name)
im.save(self.conf.out_dir + '/prediction' + filename)
#resized = Image.fromarray(resized_dec[0], mode='RGB')
#fn = '/%s_resized_dec.png' % (img_name)
#resized.save(self.conf.out_dir + '/resized_decoding' + fn)
# Save predictions for visualization.
# See utils/label_utils.py for color setting
# Need to be modified based on datasets.
if self.conf.visual:
msk = decode_labels(preds, num_classes=self.conf.num_classes)
im = Image.fromarray(msk[0], mode='RGB')
filename = '/%s_mask_visual.png' % (img_name)
im.save(self.conf.out_dir + '/visual_prediction' + filename)
if step % 100 == 0:
print('step {:d}'.format(step))
print('The output files has been saved to {}'.format(
self.conf.out_dir))
# finish
self.coord.request_stop()
self.coord.join(threads)
def predict(self):
self.predict_setup()
self.sess.run([tf.global_variables_initializer(), tf.local_variables_initializer()])
# checkpoing_file
checkpoint_file = tf.train.latest_checkpoint(self.conf.modeldir)
if (not os.path.exists("{}.meta".format(checkpoint_file))) and (self.conf.pretrain_file is not None):
self.load(self.loader, self.conf.pretrain_file)
elif os.path.exists("{}.meta".format(checkpoint_file)):
self.load(self.loader, checkpoint_file)
# Start queue threads.
threads = tf.train.start_queue_runners(coord=self.coord, sess=self.sess)
# img_name_list
image_list, _ = read_labeled_image_list('', self.conf.test_data_list)
# Predict!
for step in range(self.conf.test_num_steps):
preds = self.sess.run(self.pred)
img_name = image_list[step].split('/')[2].split('.')[0]
# Save raw predictions, i.e. each pixel is an integer between [0,20].
im = Image.fromarray(preds[0, :, :, 0], mode='L')
filename = '/%s_mask.png' % (img_name)
im.save(self.conf.out_dir + '/prediction' + filename)
# Save predictions for visualization.
if self.conf.visual:
msk = decode_labels(preds, num_classes=self.conf.num_classes)
im = Image.fromarray(msk[0], mode='RGB')
im.save(self.conf.out_dir + '/visual_prediction' + '/{}_mask_visual.png'.format(img_name))
# 原图
origin_image = "{}{}".format(self.conf.data_dir, image_list[step])
im = Image.open(origin_image)
filename, ext = os.path.splitext(filename)
im.save(self.conf.out_dir + '/visual_prediction' + filename + "_original" + ext)
if step % 100 == 0:
print('step {:d}'.format(step))
print('The output files has been saved to {}'.format(self.conf.out_dir))
# finish
self.coord.request_stop()
self.coord.join(threads)
pass
def predict(self):
self.predict_setup()
self.sess.run([
tf.local_variables_initializer(),
tf.global_variables_initializer()
])
# load checkpoint
self.load(
self.loader,
self.conf.modeldir + '/model.ckpt-' + str(self.conf.valid_step))
# Start queue threads.
threads = tf.train.start_queue_runners(coord=self.coord,
sess=self.sess)
# img_name_list
image_list, _ = read_labeled_image_list('', self.conf.test_data_list)
# Predict!
for step in range(self.conf.test_num_steps):
preds = self.sess.run(self.pred)
img_name = image_list[step].split('/')[2].split('.')[0]
# Save raw predictions, i.e. each pixel is an integer between [0,20].
im = Image.fromarray(preds[0, :, :, 0], mode='L')
im.save(self.conf.out_dir +
'/prediction/{}_mask.png'.format(img_name))
if self.conf.visual:
msk = decode_labels(preds, num_classes=self.conf.num_classes)
im = Image.fromarray(msk[0], mode='RGB')
im.save(
self.conf.out_dir +
'/visual_prediction/{}_mask_visual.png'.format(img_name))
if step % 100 == 0:
print('step {:d}'.format(step))
print('The output files has been saved to {}'.format(
self.conf.out_dir))
# finish
self.coord.request_stop()
self.coord.join(threads)
pass
def predict(self):
self.predict_setup()
self.sess.run(tf.global_variables_initializer())
self.sess.run(tf.local_variables_initializer())
# load checkpoint
checkpointfile = self.conf.modeldir+ '/model.ckpt-' + str(self.conf.valid_step)
self.load(self.loader, checkpointfile)
# Start queue threads.
threads = tf.train.start_queue_runners(coord=self.coord, sess=self.sess)
# img_name_list
image_list, _ = read_labeled_image_list('', self.conf.test_data_list)
# Predict!
for step in range(self.conf.test_num_steps):
preds = self.sess.run(self.pred)
img_name = image_list[step].split('/')[2].split('.')[0]
# Save raw predictions, i.e. each pixel is an integer between [0,20].
im = Image.fromarray(preds[0,:,:,0], mode='L')
filename = '/%s_mask.png' % (img_name)
im.save(self.conf.out_dir + '/prediction' + filename)
# Save predictions for visualization.
# See utils/label_utils.py for color setting
# Need to be modified based on datasets.
if self.conf.visual:
msk = decode_labels(preds, num_classes=self.conf.num_classes)
im = Image.fromarray(msk[0], mode='RGB')
filename = '/%s_mask_visual.png' % (img_name)
im.save(self.conf.out_dir + '/visual_prediction' + filename)
if step % 100 == 0:
print('step {:d}'.format(step))
print('The output files has been saved to {}'.format(self.conf.out_dir))
# finish
self.coord.request_stop()
self.coord.join(threads)
def predict(self):
self.predict_setup()
self.sess.run(tf.global_variables_initializer())
self.sess.run(tf.local_variables_initializer())
# load checkpoint
checkpointfile = self.conf.modeldir + '/model.ckpt-' + str(
self.conf.valid_step)
self.load(self.loader, checkpointfile)
# Start queue threads.
threads = tf.train.start_queue_runners(coord=self.coord,
sess=self.sess)
# img_name_list
image_list, _ = read_labeled_image_list('', self.conf.test_data_list)
# Predict!
for step in range(self.conf.test_num_steps):
preds = self.sess.run(self.pred)
predmatrix = preds[0, :, :, 0, :]
check = np.sum(predmatrix, axis=2)
groesse = predmatrix.shape
inhalt = ""
zeile = ""
datei = "../superpixel.zip"
img_name = image_list[step].split('/')[2].split('.')[0]
preddict = {}
countdict = {}
meanpred = {}
with zipfile.ZipFile(datei) as z:
inhalt = z.open(img_name + ".csv")
spdf = pd.read_csv(inhalt, header=None)
sparray = spdf.values
z.close()
if not (sparray.shape[0] == groesse[0]
and sparray.shape[1] == groesse[1]):
print("ERROR WITH IMAGE: ", img_name,
"\tThey dont have the same size")
continue
for x in range(groesse[0]):
for y in range(groesse[1]):
superpixel = sparray[x, y]
#initialize dict:
if superpixel not in preddict.keys():
preddict[superpixel] = [0.0] * self.conf.num_classes
countdict[superpixel] = 0
meanpred[superpixel] = [0.0] * self.conf.num_classes
countdict[superpixel] += 1
for klasse in range(groesse[2]):
preddict[superpixel][klasse] += predmatrix[x, y,
klasse]
towrite = ""
spliste = list(preddict.keys())
spliste = sorted(spliste)
for sp in spliste:
towrite += str(sp)
for k in range(self.conf.num_classes):
meanpred[sp][k] = round(preddict[sp][k] / countdict[sp], 3)
towrite += "," + str(meanpred[sp][k])
towrite += "\n"
with open(
self.conf.out_dir + '/visual_prediction/probsp_eins/' +
img_name + ".csv", 'w') as f:
f.write(towrite)
if step < 100:
print("saving: ", img_name)
# for i in range(self.conf.num_classes):
# img_name = image_list[step].split('/')[2].split('.')[0]
# # Save raw predictions, i.e. each pixel is an integer between [0,20].
# im = Image.fromarray(preds[0, :, :, 0, i], mode='L')
# if not os.path.isdir(self.conf.out_dir + "/prediction/" + str(i)):
# os.mkdir(self.conf.out_dir + "/prediction/" + str(i))
# filename = '/%s_mask.png' % (img_name)
# im.save(self.conf.out_dir + '/prediction/' + str(i) + filename)
#
# # Save predictions for visualization.
# # See utils/label_utils.py for color setting
# # Need to be modified based on datasets.
# if self.conf.visual:
# #msk = decode_labels(preds[:, :, :, :, i], num_classes=self.conf.num_classes) #old version
# #im = Image.fromarray(msk[0], mode='RGB') #old version
# if not os.path.isdir(self.conf.out_dir + '/visual_prediction/' + str(i)):
# os.mkdir(self.conf.out_dir + '/visual_prediction/' + str(i))
# filename = '/%s_mask_visual.png' % (img_name)
# print("saving Image")
# #im.save(self.conf.out_dir + '/visual_prediction/' + str(i) + filename) #old version
# plt.imsave(self.conf.out_dir + '/visual_prediction/' + str(i) + filename, preds[0, :, :, 0, i])
if step % 100 == 0:
print('step {:d}'.format(step))
print('The output files has been saved to {}'.format(
self.conf.out_dir))
# finish
self.coord.request_stop()
self.coord.join(threads)
def predict(self):
self.predict_setup()
self.sess.run(tf.global_variables_initializer())
self.sess.run(tf.local_variables_initializer())
# load checkpoint
checkpointfile = self.conf.modeldir + '/model.ckpt-' + str(
self.conf.valid_step)
self.load(self.loader, checkpointfile)
# Start queue threads.
threads = tf.train.start_queue_runners(coord=self.coord,
sess=self.sess)
# img_name_list
image_list, _ = read_labeled_image_list('', self.conf.test_data_list)
# Predict!
for step in range(self.conf.test_num_steps):
preds = self.sess.run(self.pred)
predmatrix = preds[0, :, :, 0, :]
check = np.sum(predmatrix, axis=2)
groesse = predmatrix.shape
inhalt = ""
zeile = ""
for x in range(groesse[0]):
for y in range(groesse[1]):
zeile = str(x) + "," + str(y)
for klasse in range(groesse[2]):
zeile += "," + str(predmatrix[x, y, klasse])
inhalt += zeile + "\n"
img_name = image_list[step].split('/')[2].split('.')[0]
with open(
self.conf.out_dir + '/visual_prediction/probcsv_eins/' +
img_name + ".csv", 'w') as f:
f.write(inhalt)
if step < 100:
print("saving: ", img_name)
# for i in range(self.conf.num_classes):
# img_name = image_list[step].split('/')[2].split('.')[0]
# # Save raw predictions, i.e. each pixel is an integer between [0,20].
# im = Image.fromarray(preds[0, :, :, 0, i], mode='L')
# if not os.path.isdir(self.conf.out_dir + "/prediction/" + str(i)):
# os.mkdir(self.conf.out_dir + "/prediction/" + str(i))
# filename = '/%s_mask.png' % (img_name)
# im.save(self.conf.out_dir + '/prediction/' + str(i) + filename)
#
# # Save predictions for visualization.
# # See utils/label_utils.py for color setting
# # Need to be modified based on datasets.
# if self.conf.visual:
# #msk = decode_labels(preds[:, :, :, :, i], num_classes=self.conf.num_classes) #old version
# #im = Image.fromarray(msk[0], mode='RGB') #old version
# if not os.path.isdir(self.conf.out_dir + '/visual_prediction/' + str(i)):
# os.mkdir(self.conf.out_dir + '/visual_prediction/' + str(i))
# filename = '/%s_mask_visual.png' % (img_name)
# print("saving Image")
# #im.save(self.conf.out_dir + '/visual_prediction/' + str(i) + filename) #old version
# plt.imsave(self.conf.out_dir + '/visual_prediction/' + str(i) + filename, preds[0, :, :, 0, i])
if step % 100 == 0:
print('step {:d}'.format(step))
print('The output files has been saved to {}'.format(
self.conf.out_dir))
# finish
self.coord.request_stop()
self.coord.join(threads)
inp_h = json_dict["input_h"]
asp = inp_w / inp_h
with tf.Session() as sess:
print("load graph")
with tf.gfile.GFile(model_path, 'rb') as f:
graph_def = tf.GraphDef()
graph_def.ParseFromString(f.read())
sess.graph.as_default()
tf.import_graph_def(graph_def, name='')
input = sess.graph.get_tensor_by_name(input)
output = sess.graph.get_tensor_by_name(output)
image_list, label_list = read_labeled_image_list(
data_path, eval_list_path)
true_pos = np.zeros(num_classes)
false_pos = np.zeros(num_classes)
false_neg = np.zeros(num_classes)
IoU = [0] * num_classes
t_sum = 0
for i, img_path in enumerate(image_list):
img = cv2.imread(img_path)
label = cv2.imread(label_list[i])
img, label = resize_with_padding(img, label,
asp) # ///ресайз и паддинг\\\
one_ch_label = label[:, :, 0]
t = timer()
# with tf.Session() as sess: