def get_classes(self):
try:
label_map_string = ''
print("getting " + self.label_file + ' from ' + self.bucket)
obj = self.s3.get_object(Bucket=self.bucket, Key=self.label_file)
label_map_string = obj['Body'].read().decode('utf-8')
print("got " + self.label_file)
self.label_map = {}
self.label_map = labelmap.StringIntLabelMap()
self.labelmap_dict = {}
text_format.Merge(label_map_string, self.label_map)
self.categories = []
for item in self.label_map.item:
self.labelmap_dict[item.id] = item.name
self.categories.append({'id': item.id, 'name': item.name})
self.category_index = label_map_util.create_category_index(
self.categories)
except botocore.exceptions.ClientError as e:
if e.response['Error']['Code'] == "NoSuchKey":
self.labelmap_dict = {}
print(self.label_file + ' was not found')
else:
print(e)
raise ValueError('Error getting ' + self.label_file + ' in ' +
self.bucket)
def _save_label_map_dict(label_map_dict, label_map_path):
label_map = string_int_label_map_pb2.StringIntLabelMap()
items = sorted(label_map_dict.items(), key=lambda x: int(x[1]))
for name, id_ in items:
item = label_map.item.add()
item.id = id_
item.name = name
_save_label_map(label_map, label_map_path)
def load_labelmap(path):
with open(path, "r") as labels_file:
label_map_string = labels_file.read()
label_map = string_int_label_map_pb2.StringIntLabelMap()
try:
text_format.Merge(label_map_string, label_map)
except text_format.ParseError:
label_map.ParseFromString(label_map_string)
return label_map
def load_labelmap(path):
with tf.gfile.GFile(path, 'r') as fid:
label_map_string = fid.read()
label_map = string_int_label_map_pb2.StringIntLabelMap()
try:
text_format.Merge(label_map_string, label_map)
except text_format.ParseError:
label_map.ParseFromString(label_map_string)
_validate_label_map(label_map)
return label_map
def main(_):
s = open('mscoco_complete_label_map.pbtxt', 'r').read()
mymap = labelmap.StringIntLabelMap()
global _label_map
_label_map = text_format.Parse(s, mymap)
channel = grpc.insecure_channel(FLAGS.server)
stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
run_num = 20
image, org = decode_image_opencv(FLAGS.image)
image = image.astype(np.uint8)
request_array = []
# batch_size_array = [1, 1, 4, 4, 1, 4]
batch_size_array = [1, 2, 4, 8, 16, 32, 64, 128]
# batch_size_array = [32, 16, 8, 4, 2, 1]
# batch_size_array = [10, 9, 8, 7, 6, 5]
for batch_size in batch_size_array:
inputs = image
for i in range(batch_size - 1):
inputs = np.append(inputs, image, axis = 0)
request = predict_pb2.PredictRequest()
request.model_spec.name = model_name
request.model_spec.signature_name = 'serving_default'
request.inputs['inputs'].CopyFrom(tf.contrib.util.make_tensor_proto(inputs, shape=inputs.shape))
request_array.append(request)
print("Done with input preparation")
raw_input("Press Enter to continue...\n")
for i in range(len(batch_size_array)):
batch_size = batch_size_array[i]
request = request_array[i]
durationSum = 0.0
for j in range(run_num):
start = time.time()
result = stub.Predict(request, 10.0)
end = time.time()
duration = end - start
print("duration = %f" % duration)
if (j != 0):
durationSum += duration
print("average duration (warm-up excluded) for batch size of %d = %f" % (batch_size, durationSum / (run_num - 1)))
def main(_):
s = open('mscoco_complete_label_map.pbtxt', 'r').read()
mymap = labelmap.StringIntLabelMap()
global _label_map
_label_map = text_format.Parse(s, mymap)
channel = grpc.insecure_channel(FLAGS.server)
stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
# durationSum = 0.0
thread_num = 1
run_num = 100
batch_size = 1
image, org = decode_image_opencv(FLAGS.image)
image = image.astype(np.uint8)
inputs = image
for i in range(batch_size - 1):
inputs = np.append(inputs, image, axis = 0)
# request = predict_pb2.PredictRequest()
# request.model_spec.name = model_name
# request.model_spec.signature_name = 'serving_default'
# request.inputs['inputs'].CopyFrom(tf.contrib.util.make_tensor_proto(inputs, shape=inputs.shape))
# # warmup
# warmup_num = 3
# for i in range(warmup_num):
# start = time.time()
# result = stub.Predict(request, 10.0)
# end = time.time()
# duration = end - start
# print("warmup duration = %f" % duration)
# # time.sleep(2.0)
raw_input("Press Enter to continue...")
start = time.time()
jobs = []
for i in range(thread_num):
p = multiprocessing.Process(target = send_request, args = (inputs, run_num, batch_size, i,))
jobs.append(p)
p.start()
for p in jobs:
p.join()
end = time.time()
print("overall time = %f" % (end - start))
def load_labelmap(path):
"""Loads label map proto.
Args:
path: path to StringIntLabelMap proto text file.
Returns:
a StringIntLabelMapProto
"""
with tf.gfile.GFile(path, 'r') as fid:
label_map_string = fid.read()
label_map = string_int_label_map_pb2.StringIntLabelMap()
try:
text_format.Merge(label_map_string, label_map)
except text_format.ParseError:
label_map.ParseFromString(label_map_string)
return label_map
def main(_):
if not FLAGS.img_path:
print('Please specify img_path -img_path=...')
return
if not FLAGS.num_tests:
print('Please specify num_tests -num_tests=n')
return
if not FLAGS.server:
print('please specify server -server host:port')
return
print("Number of test=", FLAGS.num_tests)
s = open('mscoco_complete_label_map.pbtxt', 'r').read()
mymap = labelmap.StringIntLabelMap()
global _label_map
_label_map = text_format.Parse(s, mymap)
do_inference(FLAGS.server, FLAGS.batch_size, FLAGS.num_tests,
FLAGS.img_path)
def cats_to_pbtxt(cats_json, pbtxt):
cats, _ = load_cats(cats_json)
cats[0] = 'none_of_the_above' #insert bg category
import sys
sys.path.append('/home/splendor/win/learn/models/object_detection/protos')
import string_int_label_map_pb2
labels = string_int_label_map_pb2.StringIntLabelMap()
for k, v in cats.items():
label = labels.item.add()
label.id = k
label.name = v
# print(labels)
with open(pbtxt, 'w') as f:
f.write(str(labels))
def _load_label_map(label_map_path):
"""Loads label map proto.
Args:
label_map_path: path to StringIntLabelMap proto text file.
Returns:
a StringIntLabelMapProto
"""
with tf.gfile.GFile(label_map_path, 'r') as fid:
label_map_string = fid.read()
label_map = string_int_label_map_pb2.StringIntLabelMap()
try:
text_format.Merge(label_map_string, label_map)
except text_format.ParseError:
label_map.ParseFromString(label_map_string)
# validate label map
for item in label_map.item:
if item.id < 1:
raise ValueError('Label map ids should be >= 1.')
return label_map
def main(_):
label_map = labelmap.StringIntLabelMap()
with open(FLAGS.label_file, 'r') as fid:
label_map_string = fid.read()
text_format.Merge(label_map_string, label_map)
categories = []
for item in label_map.item:
labelmap_dict[item.name] = item.id
categories.append({'id': item.id, 'name': item.name})
print(categories)
print(labelmap_dict)
writer = tf.python_io.TFRecordWriter(FLAGS.output_path)
path = os.path.join(os.getcwd(), FLAGS.image_dir)
examples = pd.read_csv(FLAGS.csv_input)
grouped = split(examples, 'filename')
for group in grouped:
tf_example = create_tf_example(group, path)
writer.write(tf_example.SerializeToString())
writer.close()
output_path = os.path.join(os.getcwd(), FLAGS.output_path)
print('Successfully created the TFRecords: {}'.format(output_path))
def load_pbtxt_file(path):
"""Read .pbtxt file.
Args:
path: Path to StringIntLabelMap proto text file (.pbtxt file).
Returns:
A StringIntLabelMapProto.
Raises:
ValueError: If path is not exist.
"""
if not tf.gfile.Exists(path):
raise ValueError('`path` is not exist.')
with tf.gfile.GFile(path, 'r') as fid:
pbtxt_string = fid.read()
pbtxt = string_int_label_map_pb2.StringIntLabelMap()
try:
text_format.Merge(pbtxt_string, pbtxt)
except text_format.ParseError:
pbtxt.ParseFromString(pbtxt_string)
return pbtxt
def main(_):
s = open('mscoco_complete_label_map.pbtxt', 'r').read()
mymap = labelmap.StringIntLabelMap()
global _label_map
_label_map = text_format.Parse(s, mymap)
channel = grpc.insecure_channel(FLAGS.server)
stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
image, org = decode_image_opencv(FLAGS.image)
image = image.astype(np.uint8)
model_1 = 'ssd_inception_v2_coco'
model_2 = 'ssd_mobilenet'
# thread_pool = []
t1 = threading.Thread(target=send_request,
args=(
stub,
image,
model_1,
10,
))
t2 = threading.Thread(target=send_request,
args=(
stub,
image,
model_2,
1,
))
t1.start()
t2.start()
t1.join()
t2.join()
def Setup():
s = open('%s/modules_traffic/mscoco_complete_label_map.pbtxt' % os.environ['TRAFFIC_JAMMER_PATH'], 'r').read()
mymap = labelmap.StringIntLabelMap()
TrafficInception._label_map = text_format.Parse(s, mymap)
def main(_):
if (sys.argv[1] == "mobilenet"):
model_name = "ssd_mobilenet_v1_coco"
elif (sys.argv[1] == "inception"):
model_name = "ssd_inception_v2_coco"
else:
model_name = "ssd_resnet50_v1_fpn"
s = open('/home/yitao/Documents/fun-project/tensorflow-related/traffic-jammer/single_dnn_client/obj_det/mscoco_complete_label_map.pbtxt','r').read()
mymap =labelmap.StringIntLabelMap()
global _label_map
_label_map = text_format.Parse(s,mymap)
channel = grpc.insecure_channel(FLAGS.server)
stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
run_num = 10
batch_size = 1
for i in range(run_num):
start = time.time()
image, org = decode_image_opencv(FLAGS.image)
_draw = org.copy()
image = image.astype(np.uint8)
inputs = image
for i in range(batch_size - 1):
inputs = np.append(inputs, image, axis = 0)
request = predict_pb2.PredictRequest()
request.model_spec.name = model_name
request.model_spec.signature_name = 'serving_default'
request.inputs['inputs'].CopyFrom(tf.contrib.util.make_tensor_proto
(inputs, shape=inputs.shape))
result = stub.Predict(request, 10.0)
# print(result)
boxes = result.outputs['detection_boxes']
scores = result.outputs['detection_scores']
labels = result.outputs['detection_classes']
num_detections= result.outputs['num_detections']
# print("???")
# print(boxes)
boxes= tf.make_ndarray(boxes)
scores= tf.make_ndarray(scores)
labels= tf.make_ndarray(labels)
num_detections= tf.make_ndarray(num_detections)
# print("boxes output",(boxes).shape)
# print("scores output",(scores).shape)
# print("labels output",(labels).shape)
# print('num_detections',num_detections[0])
# # visualize detections hints from
# # # https://github.com/tensorflow/models/blob/master/research/object_detection/object_detection_tutorial.ipynb
# for box, score, label in zip(boxes[0], scores[0], labels[0]):
# # scores are sorted so we can break
# if score < 0.3:
# break
# #dim = image.shape[0:2]
# dim = _draw.shape
# #print("Label-raw",labels_to_names[label-1]," at ",box," Score ",score)
# box = box_normal_to_pixel(box, dim)
# b = box.astype(int)
# class_label = get_label(int(label))
# print("Label",class_label ," at ",b," Score ",score)
# # draw the image and write out
# cv2.rectangle(_draw,(b[0],b[1]),(b[2],b[3]),(0,0,255),1)
# cv2.putText(_draw,class_label + "-"+str(round(score,2)), (b[0]+2,b[1]+8),\
# cv2.FONT_HERSHEY_SIMPLEX, .45, (0,0,255))
# cv2.imshow("test", _draw)
# cv2.waitKey(0)
end = time.time()
duration = end - start
print("duration = %s sec" % str(duration))
def generate_elements(self, url):
train_writer = tf.python_io.TFRecordWriter(dir_path + 'train.record')
test_writer = tf.python_io.TFRecordWriter(dir_path + 'test.record')
try:
if os.path.isfile(labelmap_file) == True:
label_file = util_obj.open_file(labelmap_file, "rb")
class_map = string_int_label_map_pb2.StringIntLabelMap()
class_map.ParseFromString(label_file.read())
label_file.close()
else:
class_map = string_int_label_map_pb2.StringIntLabelMap()
self.driver.get(url)
#all_visible_elements = self.driver.find_elements_by_xpath("//*[not(contains(@style,'display:none'))]")
all_visible_elements = self.driver.find_elements_by_xpath("//a[not(contains(@style,'display:none')) and not(contains(@class,'story'))] | //h2[not(contains(@style,'display:none')) and not(contains(@class,'story'))] | //button[not(contains(@style,'display:none')) and not(contains(@class,'story'))] | //span[not(contains(@style,'display:none')) and not(contains(@class,'story'))] | //input[not(contains(@style,'display:none')) and not(contains(@class,'story'))]")
for element in all_visible_elements:
elements_list = []
location = element.location_once_scrolled_into_view
location = element.location
size = element.size
x1_orig = location['x'] * 2
y1_orig = location['y'] * 2
x2_orig = x1_orig + size['width'] * 2
y2_orig = y1_orig + size['height'] * 2
words = element.text.split()
if x2_orig != x1_orig and y2_orig != y1_orig and len(words) < 4:
png = self.driver.get_screenshot_as_png()
pil_image = Image.open(BytesIO(png))
image_width, image_height = pil_image.size
img, x1,y1,x2,y2 = self.crop_image(pil_image,x1_orig,y1_orig,x2_orig,y2_orig,1024,600)
#pil_image = pil_image.resize((1024,600))
img = np.asarray(img)
suffix = self.get_suffix_from_element(element)
#x1 = int((x1_orig/image_width)*1024)
#x2 = int((x2_orig/image_width)*1024)
#y1 = int((y1_orig/image_height)*600)
#y2 = int((y2_orig/image_height)*600)
if not suffix:
continue
class_name = element.tag_name + '_' + suffix
original_image_path = dir_path + "images/train/" + class_name + '.png'
#pil_image.save(original_image_path)
self.image_with_bounding_box(samples_dir,class_name,img,x1,y1,x2,y2)
# writing class to label_map.pbtxt
class_id = util_obj.add_class_to_label_map(class_name, class_map)
print("class id", class_id)
# creating an tf.example
#width = size['width'] * 2
#height = size['height'] * 2
elements_list.append([img,x1,y1,x2,y2,image_width,image_height,class_name,class_id])
train_list,test_list = util_obj.create_augmented_images(elements_list,80)
self.serialize_image_list(train_writer,train_list)
self.serialize_image_list(test_writer,test_list)
except Exception:
#print(class_name)
traceback.print_exc()
label_map_file = util_obj.open_file(labelmap_file, "wb")
label_map_file.write(class_map.SerializeToString())
label_map_file.close()
train_writer.close()
test_writer.close()
def main(_):
s = open('mscoco_complete_label_map.pbtxt', 'r').read()
mymap = labelmap.StringIntLabelMap()
global _label_map
_label_map = text_format.Parse(s, mymap)
channel = grpc.insecure_channel(FLAGS.server)
stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
thread_num_array = [1, 2, 4]
batch_size_2d_array = [[1, 2, 4, 8, 16, 32, 64, 128],
[1, 2, 4, 8, 16, 32, 64], [1, 2, 4, 8, 16, 32]]
# thread_num_array = [1]
# batch_size_2d_array = [[4, 8]]
for i in range(len(thread_num_array)):
thread_num = thread_num_array[i]
batch_size_array = batch_size_2d_array[i]
for batch_size in batch_size_array:
time.sleep(2.0)
run_num = 1024 / (thread_num * batch_size)
# print("[INFO] thread_num = %d, batch_size = %d, run_num = %d" % (thread_num, batch_size, run_num))
t0 = time.time()
inputs = prepareInputs(batch_size)
t1 = time.time()
# print("[Debug] it took %.3f sec to prepare image of shape %s" % (t1 - t0, inputs.shape))
# warmup
warmup_num = 3
send_request(stub, inputs, batch_size, warmup_num, 0)
# raw_input("Press Enter to continue...")
time.sleep(2.0)
t0 = time.time()
thread_pool = []
for thread_id in range(thread_num):
t = threading.Thread(target=send_request,
args=(
stub,
inputs,
batch_size,
run_num,
thread_id,
))
thread_pool.append(t)
t.start()
for t in thread_pool:
t.join()
t1 = time.time()
print(
"[Summary] thread_num = %d, batch_size = %d, run_num = %d, total duration = %.3f"
% (thread_num, batch_size, run_num, t1 - t0))
def main(_):
s = open('mscoco_complete_label_map.pbtxt', 'r').read()
mymap = labelmap.StringIntLabelMap()
global _label_map
_label_map = text_format.Parse(s, mymap)
channel = grpc.insecure_channel(FLAGS.server)
stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
# durationSum = 0.0
thread_num = int(sys.argv[1])
# run_num = 10
batch_size = int(sys.argv[2])
run_num = 1024 / (thread_num * batch_size)
print("[INFO] thread_num = %d, batch_size = %d, run_num = %d" %
(thread_num, batch_size, run_num))
image, org = decode_image_opencv(FLAGS.image)
image = image.astype(np.uint8)
t0 = time.time()
inputs = image
for i in range(batch_size - 1):
inputs = np.append(inputs, image, axis=0)
t1 = time.time()
print("\n[INFO] preparing input data takes %.3f sec\n" % (t1 - t0))
request = predict_pb2.PredictRequest()
request.model_spec.name = model_name
request.model_spec.signature_name = 'serving_default'
request.inputs['inputs'].CopyFrom(
tf.contrib.util.make_tensor_proto(inputs, shape=inputs.shape))
# warmup
warmup_num = 3
for i in range(warmup_num):
start = time.time()
result = stub.Predict(request, 10.0)
end = time.time()
duration = end - start
print("warmup duration = %f" % duration)
# time.sleep(2.0)
raw_input("Press Enter to continue...")
start = time.time()
thread_pool = []
for i in range(thread_num):
t = threading.Thread(target=send_request,
args=(
stub,
inputs,
run_num,
batch_size,
i,
))
thread_pool.append(t)
t.start()
for t in thread_pool:
t.join()
end = time.time()
print("overall time = %f" % (end - start))
def main(_):
s = open('mscoco_complete_label_map.pbtxt', 'r').read()
mymap = labelmap.StringIntLabelMap()
global _label_map
_label_map = text_format.Parse(s, mymap)
channel = grpc.insecure_channel(FLAGS.server)
stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
durationSum = 0.0
run_num = 10
batch_size = 64
image, org = decode_image_opencv(FLAGS.image)
# _draw = org.copy()
image = image.astype(np.uint8)
inputs = image
for i in range(batch_size - 1):
inputs = np.append(inputs, image, axis=0)
request = predict_pb2.PredictRequest()
request.model_spec.name = model_name
request.model_spec.signature_name = 'serving_default'
request.inputs['inputs'].CopyFrom(
tf.contrib.util.make_tensor_proto(inputs, shape=inputs.shape))
for i in range(run_num):
start = time.time()
result = stub.Predict(request, 10.0)
end = time.time()
if (i != 0):
duration = end - start
durationSum += duration
print("duration = %f" % duration)
# print(result)
# boxes = result.outputs['detection_boxes']
# scores = result.outputs['detection_scores']
# labels = result.outputs['detection_classes']
# num_detections= result.outputs['num_detections']
# boxes= tf.make_ndarray(boxes)
# scores= tf.make_ndarray(scores)
# labels= tf.make_ndarray(labels)
# num_detections= tf.make_ndarray(num_detections)
# print("boxes output",(boxes).shape)
# print("scores output",(scores).shape)
# print("labels output",(labels).shape)
# print('num_detections',num_detections[0])
# # visualize detections hints from
# # # https://github.com/tensorflow/models/blob/master/research/object_detection/object_detection_tutorial.ipynb
# for box, score, label in zip(boxes[0], scores[0], labels[0]):
# # scores are sorted so we can break
# if score < 0.3:
# break
# #dim = image.shape[0:2]
# dim = _draw.shape
# #print("Label-raw",labels_to_names[label-1]," at ",box," Score ",score)
# box = box_normal_to_pixel(box, dim)
# b = box.astype(int)
# class_label = get_label(int(label))
# print("Label",class_label ," at ",b," Score ",score)
# # draw the image and write out
# cv2.rectangle(_draw,(b[0],b[1]),(b[2],b[3]),(0,0,255),1)
# cv2.putText(_draw,class_label + "-"+str(round(score,2)), (b[0]+2,b[1]+8),\
# cv2.FONT_HERSHEY_SIMPLEX, .45, (0,0,255))
# cv2.imshow("test", _draw)
# cv2.waitKey(0)
print("average duration for batch size of %d = %f" %
(batch_size, durationSum / (run_num - 1)))