1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
|
"""
Copyright (C) 2018-2019 Intel Corporation
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
"""
from datetime import datetime
from statistics import median
from openvino.inference_engine import IENetwork, IECore, get_version
from .utils.constants import CPU_DEVICE_NAME, MULTI_DEVICE_NAME, GPU_DEVICE_NAME, MYRIAD_DEVICE_NAME
from .utils.logging import logger
from .utils.utils import get_duration_seconds, parse_value_per_device, parse_devices
class Benchmark:
def __init__(self, device: str, number_infer_requests, number_iterations, duration_seconds, api_type):
self.device = device.upper()
self.ie = IECore()
self.nireq = number_infer_requests
self.niter = number_iterations
self.duration_seconds = get_duration_seconds(duration_seconds, self.niter, self.device)
self.api_type = api_type
self.device_number_streams = {}
def __del__(self):
del self.ie
def add_extension(self, path_to_extension: str=None, path_to_cldnn_config: str=None):
if GPU_DEVICE_NAME in self.device:
if path_to_cldnn_config:
self.ie.set_config({'CONFIG_FILE': path_to_cldnn_config}, GPU_DEVICE_NAME)
logger.info('GPU extensions is loaded {}'.format(path_to_cldnn_config))
if CPU_DEVICE_NAME in self.device or MYRIAD_DEVICE_NAME in self.device:
if path_to_extension:
self.ie.add_extension(extension_path=path_to_extension, device_name=CPU_DEVICE_NAME)
logger.info('CPU extensions is loaded {}'.format(path_to_extension))
def get_version_info(self) -> str:
logger.info('InferenceEngine:\n{: <9}{:.<24} {}'.format('', 'API version', get_version()))
version_string = 'Device info\n'
for device, version in self.ie.get_versions(self.device).items():
version_string += '{: <9}{}\n'.format('', device)
version_string += '{: <9}{:.<24}{} {}.{}\n'.format('', version.description, ' version', version.major,
version.minor)
version_string += '{: <9}{:.<24} {}\n'.format('', 'Build', version.build_number)
return version_string
@staticmethod
def reshape(ie_network: IENetwork, batch_size: int):
new_shapes = {}
for input_layer_name, input_layer in ie_network.inputs.items():
shape = input_layer.shape
layout = input_layer.layout
try:
batch_index = layout.index('N')
except ValueError:
batch_index = 1 if layout == 'C' else -1
if batch_index != -1 and shape[batch_index] != batch_size:
shape[batch_index] = batch_size
new_shapes[input_layer_name] = shape
if new_shapes:
logger.info('Resizing network to batch = {}'.format(batch_size))
ie_network.reshape(new_shapes)
def set_config(self, number_streams: int, api_type: str = 'async',
number_threads: int = None, infer_threads_pinning: int = None):
devices = parse_devices(self.device)
self.device_number_streams = parse_value_per_device(devices, number_streams)
for device in devices:
if device == CPU_DEVICE_NAME: # CPU supports few special performance-oriented keys
# limit threading for CPU portion of inference
if number_threads:
self.ie.set_config({'CPU_THREADS_NUM': str(number_threads)}, device)
if MULTI_DEVICE_NAME in self.device and GPU_DEVICE_NAME in self.device:
self.ie.set_config({'CPU_BIND_THREAD': 'NO'}, CPU_DEVICE_NAME)
else:
# pin threads for CPU portion of inference
self.ie.set_config({'CPU_BIND_THREAD': infer_threads_pinning}, device)
# for CPU execution, more throughput-oriented execution via streams
# for pure CPU execution, more throughput-oriented execution via streams
if api_type == 'async':
cpu_throughput = {'CPU_THROUGHPUT_STREAMS': 'CPU_THROUGHPUT_AUTO'}
if device in self.device_number_streams.keys():
cpu_throughput['CPU_THROUGHPUT_STREAMS'] = str(self.device_number_streams.get(device))
self.ie.set_config(cpu_throughput, device)
self.device_number_streams[device] = self.ie.get_config(device, 'CPU_THROUGHPUT_STREAMS')
elif device == GPU_DEVICE_NAME:
if api_type == 'async':
gpu_throughput = {'GPU_THROUGHPUT_STREAMS': 'GPU_THROUGHPUT_AUTO'}
if device in self.device_number_streams.keys():
gpu_throughput['GPU_THROUGHPUT_STREAMS'] = str(self.device_number_streams.get(device))
self.ie.set_config(gpu_throughput, device)
self.device_number_streams[device] = self.ie.get_config(device, 'GPU_THROUGHPUT_STREAMS')
if MULTI_DEVICE_NAME in self.device and CPU_DEVICE_NAME in self.device:
# multi-device execution with the CPU+GPU performs best with GPU trottling hint,
# which releases another CPU thread (that is otherwise used by the GPU driver for active polling)
self.ie.set_config({'CLDNN_PLUGIN_THROTTLE': '1'}, device)
elif device == MYRIAD_DEVICE_NAME:
self.ie.set_config({'LOG_LEVEL': 'LOG_INFO',
'VPU_LOG_LEVEL': 'LOG_WARNING'}, MYRIAD_DEVICE_NAME)
def load_network(self, ie_network: IENetwork, perf_counts: bool, number_infer_requests: int = None):
config = {'PERF_COUNT': ('YES' if perf_counts else 'NO')}
exe_network = self.ie.load_network(ie_network,
self.device,
config=config,
num_requests=number_infer_requests or 0)
return exe_network
def infer(self, request_queue, requests_input_data, batch_size, progress_bar):
progress_count = 0
# warming up - out of scope
infer_request = request_queue.get_idle_request()
if not infer_request:
raise Exception('No idle Infer Requests!')
if self.api_type == 'sync':
infer_request.infer(requests_input_data[infer_request.req_id])
else:
infer_request.start_async(requests_input_data[infer_request.req_id])
request_queue.wait_all()
request_queue.reset_times()
start_time = datetime.now()
exec_time = (datetime.now() - start_time).total_seconds()
iteration = 0
# Start inference & calculate performance
# to align number if iterations to guarantee that last infer requests are executed in the same conditions **/
while (self.niter and iteration < self.niter) or \
(self.duration_seconds and exec_time < self.duration_seconds) or \
(self.api_type == 'async' and iteration % self.nireq):
infer_request = request_queue.get_idle_request()
if not infer_request:
raise Exception('No idle Infer Requests!')
if self.api_type == 'sync':
infer_request.infer(requests_input_data[infer_request.req_id])
else:
infer_request.start_async(requests_input_data[infer_request.req_id])
iteration += 1
exec_time = (datetime.now() - start_time).total_seconds()
if self.duration_seconds:
# calculate how many progress intervals are covered by current iteration.
# depends on the current iteration time and time of each progress interval.
# Previously covered progress intervals must be skipped.
progress_interval_time = self.duration_seconds / progress_bar.total_num
new_progress = int(exec_time / progress_interval_time - progress_count)
progress_bar.add_progress(new_progress)
progress_count += new_progress
elif self.niter:
progress_bar.add_progress(1)
# wait the latest inference executions
request_queue.wait_all()
total_duration_sec = request_queue.get_duration_in_seconds()
times = request_queue.times
times.sort()
latency_ms = median(times)
fps = batch_size * 1000 / latency_ms
if self.api_type == 'async':
fps = batch_size * iteration / total_duration_sec
progress_bar.finish()
return fps, latency_ms, total_duration_sec, iteration
|
