Skip to content

GitLab

"megatron/git@developer.sourcefind.cn:wuxk1/megatron-lm.git" did not exist on "fe794c5ae848ef6e650c0ac501c68b635e592e39"
Commit 24b257f1 authored by sunzhq2's avatar sunzhq2
Browse files

init

parent 920b3c0f
Hide whitespace changes
Inline Side-by-side
Showing with 1214 additions and 0 deletions
ByteMLPerf/byte_infer_perf/general_perf/backends/runtime_backend.py 0 → 100644
View file @ 24b257f1
# Copyright 2023 ByteDance and/or its affiliates.
#
# 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 typing import Any, Dict
class RuntimeBackend(object):
def __init__(self):
self.hardware_type = 'UnKnown'
self.need_reload = False
self.need_quant = False
def version(self) -> str:
"""
Return runtime backend version details
"""
raise NotImplementedError("RuntimeBackend:version")
def load(self, batch_size) -> str:
"""
Return runtime backend version details
"""
raise NotImplementedError("RuntimeBackend:load")
def get_loaded_batch_size(self) -> int:
"""
Get Currect batch size
"""
raise NotImplementedError("RuntimeBackend:get_loaded_batch_size")
def predict(self, data):
"""
Run the compiled model and return the model output corresponding to the data.
"""
raise NotImplementedError("RuntimeBackend:predict")
def is_qs_mode_supported(self) -> bool:
"""
Used to check whether QSv2 Runtime is enabled
"""
return False
def generate_qs_config(self) -> Dict[str, Any]:
"""
Used only when is_qs_ported return True. Generate QS Config
File for QSv2 Runtime
"""
return None
def benchmark(self, dataloader):
"""
Performance Testing when qs mode is not enabled.
"""
raise NotImplementedError("RuntimeBackend:benchmark")
ByteMLPerf/byte_infer_perf/general_perf/core/configs/backend_store.py 0 → 100644
View file @ 24b257f1
# Copyright 2023 ByteDance and/or its affiliates.
#
# 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.
import importlib
import logging
from general_perf.backends.compile_backend import CompileBackend
from general_perf.backends.runtime_backend import RuntimeBackend
log = logging.getLogger("BackendStore")
__all__ = [
"CompileBackend",
]
def init_compile_backend(hardware_type: str) -> CompileBackend:
"""
Load related compile backend with input hardware type
Arguments: str
Returns: CompileBackend()
"""
log.info("Loading Compile Backend: {}".format(hardware_type))
compile_backend = importlib.import_module('general_perf.backends.' +
hardware_type +
".compile_backend_" +
hardware_type.lower())
compile_backend = getattr(compile_backend,
"CompileBackend" + hardware_type)
return compile_backend()
def init_runtime_backend(hardware_type: str) -> RuntimeBackend:
"""
Load related compile backend with input hardware type
Arguments: str
Returns: RuntimeBackend()
"""
log.info("Loading Runtime Backend: {}".format(hardware_type))
runtime_backend = importlib.import_module('general_perf.backends.' +
hardware_type +
".runtime_backend_" +
hardware_type.lower())
runtime_backend = getattr(runtime_backend,
"RuntimeBackend" + hardware_type)
return runtime_backend()
ByteMLPerf/byte_infer_perf/general_perf/core/configs/dataset_store.py 0 → 100644
View file @ 24b257f1
# Copyright 2023 ByteDance and/or its affiliates.
#
# 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.
import importlib
import logging
from typing import Any, Dict
import os
import sys
from general_perf.datasets.data_loader import Dataset
log = logging.getLogger("DatasetStore")
def load_dataset(config: Dict[str, Any]) -> Dataset:
"""
Load related dataset class with config file
Args: Dict
Returns: Dataloader()
"""
if config['dataset_name']:
dataset_name = config['dataset_name']
log.info("Loading Dataset: {}".format(config['dataset_name']))
else:
dataset_name = 'fake_dataset'
log.info("Loading Dataset: Dataset does not exist, using fake data")
data_loader = importlib.import_module('general_perf.datasets.' +
dataset_name + ".data_loader")
data_loader = getattr(data_loader, 'DataLoader')
dataset = data_loader(config)
return dataset
ByteMLPerf/byte_infer_perf/general_perf/core/configs/workload_store.py 0 → 100644
View file @ 24b257f1
# Copyright 2023 ByteDance and/or its affiliates.
#
# 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.
import json
import os
import logging
from typing import Any, List, Dict
log = logging.getLogger("WorkloadStore")
def load_workload(task: str) -> Dict[str, Any]:
"""
Return a list of dictionary with model Configuration
Args: List[str]
Returns: List[dic]
"""
modules_dir = os.path.dirname(os.path.dirname(
os.path.dirname(__file__))) + '/workloads'
for file in os.listdir(modules_dir):
path = os.path.join(modules_dir, file)
if (not file.startswith('_') and not file.startswith('.')
and (file.endswith('.json') or os.path.isdir(path))
and file[:file.find('.json')] == task):
module_name = file
with open("general_perf/workloads/" + module_name, 'r') as f:
workload_dict = json.load(f)
return workload_dict
else:
log.error(
"Task name: [ {} ] was not found, please check your task name".
format(task))
\ No newline at end of file
ByteMLPerf/byte_infer_perf/general_perf/core/perf_engine.py 0 → 100644
View file @ 24b257f1
# Copyright 2023 ByteDance and/or its affiliates.
#
# 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.
import sys
import os
import logging
import importlib
import json
import subprocess
import time
import traceback
from typing import Any, Dict, Tuple
import virtualenv
from prompt_toolkit.shortcuts import radiolist_dialog, input_dialog, yes_no_dialog
from prompt_toolkit.styles import Style
BYTE_MLPERF_ROOT = os.path.dirname(
os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
os.chdir(BYTE_MLPERF_ROOT)
sys.path.insert(0, BYTE_MLPERF_ROOT)
import argparse
from general_perf.core.configs.workload_store import load_workload
from general_perf.core.configs.dataset_store import load_dataset
from general_perf.core.configs.backend_store import init_compile_backend, init_runtime_backend
from general_perf.tools.build_pdf import build_pdf
logging.basicConfig(level=logging.INFO)
log = logging.getLogger("PerfEngine")
os.environ["TF_CPP_MIN_LOG_LEVEL"] = '3'
def get_args():
"""Parse commandline."""
parser = argparse.ArgumentParser()
parser.add_argument(
"--task",
default="resnet50-tf-fp32",
help="The task going to be evaluted, refs to workloads/")
parser.add_argument(
"--hardware_type",
default="GPU",
help="The backend going to be evaluted, refs to backends/")
parser.add_argument("--compile_only",
action='store_true',
help="Run compilation only")
args = parser.parse_args()
return args
class PerfEngine:
def __init__(self) -> None:
super().__init__()
self.args = get_args()
self.workload = load_workload(self.args.task)
self.backend_type = self.args.hardware_type
self.compile_backend = None
self.old_os_path = os.environ['PATH']
self.prev_sys_path = list(sys.path)
self.real_prefix = sys.prefix
self.compile_only_mode = False
self.version = self.get_version()
def get_version(self):
version = ""
try:
version_file = os.path.join(str(BYTE_MLPERF_ROOT), "../VERSION")
with open(version_file) as f:
_version = f.read().splitlines()
version = '.'.join(v.split('=')[1] for v in _version)
except Exception as e:
traceback.print_exc()
log.warning(f"get bytemlperf version failed, error msg: {e}")
return version
def start_engine(self) -> None:
'''
Byte MlPerf will create an virtual env for each backend to avoid dependance conflict
'''
success, total = 0, len(self.workload)
if total == 0:
return
log.info("******************* Backend Env Initization *******************")
status = self.activate_venv(self.backend_type)
if not status:
log.warning("Activate virtualenv Failed, Please Check...")
self.compile_backend = init_compile_backend(self.backend_type)
self.runtime_backend = init_runtime_backend(self.backend_type)
output_dir = os.path.abspath('general_perf/reports/' +
self.backend_type)
os.makedirs(output_dir, exist_ok=True)
status = self.single_workload_perf(self.workload)
def single_workload_perf(
self, workload: Dict[str, Any]) -> bool:
log.info("******************************************* Start to test model: {}. *******************************************".format(workload['model']))
# Check Compile Only Mode
self.compile_only_mode = False
if self.args.compile_only or workload['compile_only']:
self.compile_only_mode = True
base_report = {
"Model": workload['model'].upper(),
"Backend": self.backend_type,
"Host Info": self.get_cpu_name()
}
# Initalize Model Config Info
model_info = self.get_model_info(workload['model'])
pre_compile_config = {"workload": workload, 'model_info': model_info}
interact_info = self.check_interact_info(pre_compile_config)
pre_compile_config['interact_info'] = interact_info
if not model_info['dataset_name']:
model_info['dataset_name'] = 'fake_dataset'
'''
Compile Backend could do some optimization like convert model format here
'''
log.info("******************************************* Running Backend Compilation... *******************************************")
log.info("Running Backend Preoptimization...")
pre_compile_config = self.compile_backend.pre_optimize(pre_compile_config)
# Initalize dataset
dataset = load_dataset(model_info)
dataset.preprocess()
base_report['Dataset'] = model_info['dataset_name'].upper(
) if model_info['dataset_name'] else None
#Placeholder Only
segment_info = self.compile_backend.segment(pre_compile_config)
best_batch_sizes = self.compile_backend.get_best_batch_size()
if isinstance(best_batch_sizes, list):
pre_compile_config['workload'][
'batch_sizes'] = best_batch_sizes
log.info("Start to compile the model...")
start = time.time()
compile_info = self.compile_backend.compile(pre_compile_config,
dataset)
end = time.time()
graph_compile_report = {}
graph_compile_report["Compile Duration"] = round(end - start, 5)
graph_compile_report["Compile Precision"] = compile_info[
'compile_precision']
graph_compile_report["Subgraph Coverage"] = compile_info['sg_percent']
if 'optimizations' in compile_info:
graph_compile_report['Optimizations'] = compile_info['optimizations']
if 'instance_count' in compile_info:
base_report['Instance Count'] = compile_info['instance_count']
if 'device_count' in compile_info:
base_report['Device Count'] = compile_info['device_count']
base_report['Graph Compile'] = graph_compile_report
# Initalize Output Dir and Reports
output_dir = os.path.abspath('general_perf/reports/' +
self.backend_type + '/' +
workload['model'])
os.makedirs(output_dir, exist_ok=True)
# Compile only mode will stop here
if self.compile_only_mode:
base_report.pop("Backend")
return compile_info["compile_status"], base_report
base_report["Version"] = self.version
base_report["Execution Date"] = time.strftime("%Y-%m-%d %H:%M:%S")
# load runtime backend
"""
Start Here
"""
batch_sizes = pre_compile_config['workload']['batch_sizes']
self.runtime_backend.configs = compile_info
self.runtime_backend.workload = workload
self.runtime_backend.model_info = model_info
self.runtime_backend.load(workload['batch_sizes'][0])
# test accuracy
accuracy_report = {}
AccuracyChecker = self.get_accuracy_checker(
model_info['dataset_name']
if model_info['dataset_name'] else 'fake_dataset')
AccuracyChecker.runtime_backend = self.runtime_backend
AccuracyChecker.dataloader = dataset
AccuracyChecker.output_dir = output_dir
AccuracyChecker.configs = compile_info
if workload['test_accuracy']:
log.info("******************************************* Running Accuracy Checker... *******************************************")
dataset.rebatch(self.runtime_backend.get_loaded_batch_size())
accuracy_results = AccuracyChecker.calculate_acc(
workload['data_percent'])
accuracy_report['Data Percent'] = workload['data_percent']
accuracy_report.update(accuracy_results)
# test numeric
if workload['test_numeric']:
log.info("******************************************* Running Numeric Checker... *******************************************")
dataset.rebatch(self.runtime_backend.get_loaded_batch_size())
if not workload['test_accuracy']:
accuracy_results = AccuracyChecker.calculate_acc(
workload['data_percent'])
diff_results = AccuracyChecker.calculate_diff()
accuracy_report.update(diff_results)
accuracy_report['Diff Dist'] = compile_info['model'] + '-to-' + compile_info['compile_precision'].lower() + ".png"
if accuracy_report:
base_report['Accuracy'] = accuracy_report
# function to test qps and latency
if workload['test_perf']:
log.info("******************************************* Runing QPS Checker... *******************************************")
performance_reports = []
qs_status = self.runtime_backend.is_qs_mode_supported()
if qs_status:
qs_config = self.runtime_backend.generate_qs_config()
performance_reports = self.qs_benchmark(qs_config)
else:
for bs in batch_sizes:
self.runtime_backend.load(bs)
batch_reports = self.runtime_backend.benchmark(dataset)
performance_reports.append(batch_reports)
base_report['Performance'] = performance_reports
if "Instance Count" not in base_report:
log.warning("Vendors need to Add # of instances")
if "Device Count" not in base_report:
log.warning("Vendors need to Add # of devices")
# write output to json file
output_report_path = output_dir + "/result-" + compile_info['compile_precision'].lower() + ".json"
with open(output_report_path, 'w') as file:
json.dump(base_report, file, indent=4)
base_report.pop("Backend")
log.info("Testing Finish. Report is saved in path: [ {}/{} ]".
format(output_dir[output_dir.rfind('general_perf'):],
os.path.basename(output_report_path)))
build_pdf(output_report_path)
log.info("PDF Version is saved in path: [ {}/{}-TO-{}.pdf ]".format(
output_dir[output_dir.rfind('general_perf'):],
base_report['Model'],
output_report_path.split('/')[-1].split('-')[1].upper()))
return compile_info["compile_status"]
#WIP
def qs_benchmark(self, qs_config: Dict[str, Any]) -> list:
return []
def get_accuracy_checker(self, dataset_name: str):
AccuracyChecker = importlib.import_module('general_perf.datasets.' +
dataset_name +
".test_accuracy")
AccuracyChecker = getattr(AccuracyChecker, 'AccuracyChecker')
return AccuracyChecker()
def get_model_info(self, model_name: str) -> Dict[str, Any]:
with open("general_perf/model_zoo/" + model_name + '.json',
'r') as file:
model_info = json.load(file)
return model_info
def get_cpu_name(self):
command = "lscpu | grep 'Model name' | awk -F: '{print $2}'"
cpu_name = subprocess.check_output(command, shell=True)
return cpu_name.decode().strip()
def check_interact_info(
self, pre_compile_config: Dict[str, Dict]) -> Dict[str, Any]:
interact_info = self.compile_backend.get_interact_profile(
pre_compile_config)
answer = {}
if len(interact_info) == 0:
return answer
dialog_style = Style.from_dict({
'dialog': 'bg:#88b8ff',
'dialog frame.label': 'bg:#ffffff #000000',
'dialog.body': 'bg:#000000 #a0acde',
'dialog shadow': 'bg:#004aaa',
})
input_style = Style.from_dict({
'dialog': 'bg:#88b8ff',
'dialog frame.label': 'bg:#ffffff #000000',
'dialog.body': 'bg:#000000 #a0acde',
'dialog shadow': 'bg:#004aaa',
'text-area.prompt': 'bg:#ffffff',
'text-area': '#000000',
})
option = yes_no_dialog(title=self.backend_type + '编译配置',
text='[请选择]：是否进行编译后端配置:',
style=dialog_style).run()
if option:
sum_question = len(interact_info)
for i, question in enumerate(interact_info):
if question['depends']:
state = 0
for title in question['depends'].split(','):
if not answer[title]:
state = 1
if state:
continue
if question['dialog_type'] == 'Yes/No Dialog':
option = yes_no_dialog(
title=self.backend_type + '编译配置进度(' + str(i + 1) +
'/' + str(sum_question) + ')',
text="[Backend " + self.backend_type + "]: " +
question['note'],
style=dialog_style).run()
elif question['dialog_type'] == "Input Dialog":
option = input_dialog(
title=self.backend_type + '编译配置进度(' + str(i + 1) +
'/' + str(sum_question) + ')',
text="[Backend " + self.backend_type + "]: " +
question['note'],
style=input_style).run()
elif question['dialog_type'] == "Radiolist Dialog":
choice = [(i, text)
for i, text in enumerate(question['options'])]
num = radiolist_dialog(
title=self.backend_type + '编译配置进度(' + str(i + 1) +
'/' + str(sum_question) + ')',
text="[Backend " + self.backend_type + "]: " +
question['note'],
values=choice,
style=dialog_style).run()
option = question['options'][num] if num is not None else question[
'default']
answer[question['name']] = option
return answer
def activate_venv(self, hardware_type: str) -> bool:
if os.path.exists('general_perf/backends/' + hardware_type +
'/requirements.txt'):
log.info("Activating Virtual Env for " + hardware_type)
venv_dir = os.path.join("general_perf/backends",
hardware_type + "/venv")
activate_file = os.path.join(venv_dir, 'bin', 'activate_this.py')
if not os.path.exists(venv_dir):
log.info("venv not exist, Creating Virtual Env for " +
hardware_type)
if (hardware_type == "HPU"):
virtualenv.create_environment(venv_dir,True)
else:
virtualenv.create_environment(venv_dir)
exec(open(activate_file).read(), {'__file__': activate_file})
python_path = os.path.join(venv_dir, 'bin', 'python3')
subprocess.call([
python_path, '-m', 'pip', 'install', '--upgrade', 'pip', '--quiet'
])
subprocess.call([
python_path, '-m', 'pip', 'install', '-r', 'general_perf/backends/' +
hardware_type + '/requirements.txt', '-q'
])
else:
exec(open(activate_file).read(), {'__file__': activate_file})
'''
just in case install failed in pre-run.
'''
python_path = os.path.join(venv_dir, 'bin', 'python3')
subprocess.call([
python_path, '-m', 'pip', 'install', '--upgrade', 'pip', '--quiet'
])
subprocess.call([
python_path, '-m', 'pip', 'install', '-r', 'general_perf/backends/' +
hardware_type + '/requirements.txt', '-q'
])
if not hasattr(sys, 'real_prefix'):
return False
return True
return True
def deactivate_venv(self):
sys.path[:
0] = self.prev_sys_path #will also revert the added site-packages
sys.prefix = self.real_prefix
os.environ['PATH'] = self.old_os_path
if __name__ == "__main__":
engine = PerfEngine()
engine.start_engine()
ByteMLPerf/byte_infer_perf/general_perf/launch.py 0 → 100644
View file @ 24b257f1
import os
import sys
import argparse
import subprocess
import logging
import json
# ${prj_root}/byte_infer_perf
BYTE_MLPERF_ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
os.chdir(BYTE_MLPERF_ROOT)
sys.path.insert(0, BYTE_MLPERF_ROOT)
from general_perf.core.configs.workload_store import load_workload
logging.basicConfig(level=logging.INFO)
log = logging.getLogger("LANUCH")
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument(
"--task",
default="",
help="The task going to be evaluted, refs to workloads/")
parser.add_argument(
"--hardware_type",
default="CPU",
help="The backend going to be evaluted, refs to backends/")
parser.add_argument("--compile_only",
action='store_true',
help="Task will stoped after compilation finished")
parser.add_argument("--show_task_list",
action='store_true',
help="Print all task names")
parser.add_argument("--show_hardware_list",
action='store_true',
help="Print all hardware bytemlperf supported")
args = parser.parse_args()
return args
def main():
parsed_args = get_args()
if parsed_args.show_task_list:
log.info("******************* Supported Task *******************")
for file in os.listdir('general_perf/workloads'):
print(file[:-5])
if parsed_args.show_hardware_list:
log.info("***************** Supported Hardware Backend *****************")
for file in os.listdir('general_perf/backends'):
if not file.endswith('.py') and not file.startswith('_'):
print(file)
if parsed_args.task:
log.info("******************* Pip Package Installing *******************")
# subprocess.call([
# 'python3', '-m', 'pip', 'install', 'pip', '--upgrade', '--quiet'])
# subprocess.call([
# 'python3', '-m', 'pip', 'install', '-r', 'general_perf/requirements.txt', '--quiet'])
workload = load_workload(parsed_args.task)
with open("general_perf/model_zoo/" + workload['model'] + '.json', 'r') as file:
model_info = json.load(file)
if not os.path.exists(model_info['model_path']):
subprocess.call([
'bash', 'general_perf/prepare_model_and_dataset.sh',
model_info['model'], model_info['dataset_name'] or "None"])
# test numeric
if workload['test_numeric'] and not parsed_args.compile_only and not workload['compile_only']:
log.info("******************************************* Running CPU Numeric Checker... *******************************************")
subprocess.call([
'bash', 'general_perf/backends/CPU/calculate_cpu_diff.sh',
workload['model'],
str(workload['batch_sizes'][0])
])
cmd = f'python3 general_perf/core/perf_engine.py --hardware_type {parsed_args.hardware_type} --task {parsed_args.task}'
if parsed_args.compile_only:
cmd += '--compile_only'
exit_code = subprocess.call(cmd, shell=True)
sys.exit(exit_code)
if __name__ == '__main__':
main()
ByteMLPerf/byte_infer_perf/general_perf/log/bert-onnxruntime-fp16.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
INFO:PerfEngine:******************************************* Start to test model: bert-onnxruntime-fp16. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_squad
INFO:SQUAD:Initial...
INFO:SQUAD:Preprocessing...
INFO:SQUAD:Rebatching batch size to: 10833 ...
0%| | 0/1 [00:00<?, ?it/s] 100%|██████████| 1/1 [00:00<00:00, 93.58it/s]
INFO:PerfEngine:Start to compile the model...
2024-10-29 10:22:36.670387581 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-10-29 10:22:36.670406950 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:SQUAD:Rebatching batch size to: 4 ...
0%| | 0/2708 [00:00<?, ?it/s] 100%|██████████| 2708/2708 [00:00<00:00, 132671.91it/s]
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/2708 [00:00<?, ?it/s] 0%| | 1/2708 [00:00<21:28, 2.10it/s] 0%| | 9/2708 [00:00<02:14, 20.07it/s] 1%| | 18/2708 [00:00<01:13, 36.43it/s] 1%| | 27/2708 [00:00<00:54, 49.14it/s] 1%|▏ | 36/2708 [00:00<00:45, 58.36it/s] 2%|▏ | 45/2708 [00:01<00:40, 65.12it/s] 2%|▏ | 54/2708 [00:01<00:37, 70.06it/s] 2%|▏ | 63/2708 [00:01<00:35, 73.67it/s] 3%|▎ | 72/2708 [00:01<00:34, 75.88it/s] 3%|▎ | 81/2708 [00:01<00:33, 77.65it/s] 3%|▎ | 90/2708 [00:01<00:33, 78.96it/s] 4%|▎ | 99/2708 [00:01<00:32, 79.92it/s] 4%|▍ | 108/2708 [00:01<00:32, 80.53it/s] 4%|▍ | 117/2708 [00:01<00:32, 80.26it/s] 5%|▍ | 126/2708 [00:02<00:32, 80.24it/s] 5%|▍ | 135/2708 [00:02<00:32, 80.33it/s] 5%|▌ | 144/2708 [00:02<00:31, 80.77it/s] 6%|▌ | 153/2708 [00:02<00:31, 80.70it/s] 6%|▌ | 162/2708 [00:02<00:31, 81.03it/s] 6%|▋ | 171/2708 [00:02<00:31, 81.22it/s] 7%|▋ | 180/2708 [00:02<00:31, 80.73it/s] 7%|▋ | 189/2708 [00:02<00:31, 80.84it/s] 7%|▋ | 198/2708 [00:02<00:30, 81.13it/s] 8%|▊ | 207/2708 [00:03<00:30, 81.32it/s] 8%|▊ | 216/2708 [00:03<00:30, 81.27it/s] 8%|▊ | 225/2708 [00:03<00:30, 80.80it/s] 9%|▊ | 234/2708 [00:03<00:30, 80.67it/s] 9%|▉ | 243/2708 [00:03<00:30, 80.99it/s] 9%|▉ | 252/2708 [00:03<00:30, 81.18it/s] 10%|▉ | 261/2708 [00:03<00:30, 81.37it/s] 10%|▉ | 270/2708 [00:03<00:29, 81.73it/s] 10%|█ | 279/2708 [00:03<00:29, 82.01it/s] 11%|█ | 288/2708 [00:04<00:29, 81.50it/s] 11%|█ | 297/2708 [00:04<00:29, 81.85it/s] 11%|█▏ | 306/2708 [00:04<00:29, 82.13it/s] 12%|█▏ | 315/2708 [00:04<00:29, 82.19it/s] 12%|█▏ | 324/2708 [00:04<00:29, 82.07it/s] 12%|█▏ | 333/2708 [00:04<00:29, 81.59it/s] 13%|█▎ | 342/2708 [00:04<00:29, 81.14it/s] 13%|█▎ | 351/2708 [00:04<00:28, 81.34it/s] 13%|█▎ | 360/2708 [00:04<00:28, 81.54it/s] 14%|█▎ | 369/2708 [00:04<00:28, 81.58it/s] 14%|█▍ | 378/2708 [00:05<00:28, 81.51it/s] 14%|█▍ | 387/2708 [00:05<00:28, 81.78it/s] 15%|█▍ | 396/2708 [00:05<00:28, 81.55it/s] 15%|█▍ | 405/2708 [00:05<00:28, 81.62it/s] 15%|█▌ | 414/2708 [00:05<00:28, 81.73it/s] 16%|█▌ | 423/2708 [00:05<00:27, 81.89it/s] 16%|█▌ | 432/2708 [00:05<00:27, 81.73it/s] 16%|█▋ | 441/2708 [00:05<00:27, 81.98it/s] 17%|█▋ | 450/2708 [00:05<00:27, 80.99it/s] 17%|█▋ | 459/2708 [00:06<00:27, 81.30it/s] 17%|█▋ | 468/2708 [00:06<00:27, 81.43it/s] 18%|█▊ | 477/2708 [00:06<00:27, 81.58it/s] 18%|█▊ | 486/2708 [00:06<00:27, 81.81it/s] 18%|█▊ | 495/2708 [00:06<00:27, 81.89it/s] 19%|█▊ | 504/2708 [00:06<00:27, 81.35it/s] 19%|█▉ | 513/2708 [00:06<00:26, 81.55it/s] 19%|█▉ | 522/2708 [00:06<00:26, 81.84it/s] 20%|█▉ | 531/2708 [00:06<00:26, 81.82it/s] 20%|█▉ | 540/2708 [00:07<00:26, 81.59it/s] 20%|██ | 549/2708 [00:07<00:26, 81.67it/s] 21%|██ | 558/2708 [00:07<00:27, 78.50it/s] 21%|██ | 567/2708 [00:07<00:26, 79.94it/s] 21%|██▏ | 576/2708 [00:07<00:26, 80.86it/s] 22%|██▏ | 585/2708 [00:07<00:26, 81.40it/s] 22%|██▏ | 594/2708 [00:07<00:25, 81.91it/s] 22%|██▏ | 603/2708 [00:07<00:25, 82.13it/s] 23%|██▎ | 612/2708 [00:07<00:25, 82.05it/s] 23%|██▎ | 621/2708 [00:08<00:25, 82.23it/s] 23%|██▎ | 630/2708 [00:08<00:25, 82.55it/s] 24%|██▎ | 639/2708 [00:08<00:25, 82.74it/s] 24%|██▍ | 648/2708 [00:08<00:24, 82.74it/s] 24%|██▍ | 657/2708 [00:08<00:24, 82.89it/s] 25%|██▍ | 666/2708 [00:08<00:24, 81.89it/s] 25%|██▍ | 675/2708 [00:08<00:24, 82.37it/s] 25%|██▌ | 684/2708 [00:08<00:24, 82.61it/s] 26%|██▌ | 693/2708 [00:08<00:24, 82.80it/s] 26%|██▌ | 702/2708 [00:09<00:24, 82.80it/s] 26%|██▋ | 711/2708 [00:09<00:24, 82.91it/s] 27%|██▋ | 720/2708 [00:09<00:24, 82.36it/s] 27%|██▋ | 729/2708 [00:09<00:23, 82.69it/s] 27%|██▋ | 738/2708 [00:09<00:23, 82.89it/s] 28%|██▊ | 747/2708 [00:09<00:23, 82.96it/s] 28%|██▊ | 756/2708 [00:09<00:23, 83.12it/s] 28%|██▊ | 765/2708 [00:09<00:23, 83.13it/s] 29%|██▊ | 774/2708 [00:09<00:23, 82.10it/s] 29%|██▉ | 783/2708 [00:10<00:23, 82.24it/s] 29%|██▉ | 792/2708 [00:10<00:23, 82.08it/s] 30%|██▉ | 801/2708 [00:10<00:23, 82.01it/s] 30%|██▉ | 810/2708 [00:10<00:23, 82.19it/s] 30%|███ | 819/2708 [00:10<00:22, 82.24it/s] 31%|███ | 828/2708 [00:10<00:23, 81.51it/s] 31%|███ | 837/2708 [00:10<00:22, 81.79it/s] 31%|███ | 846/2708 [00:10<00:22, 81.99it/s] 32%|███▏ | 855/2708 [00:10<00:22, 82.03it/s] 32%|███▏ | 864/2708 [00:11<00:22, 82.14it/s] 32%|███▏ | 873/2708 [00:11<00:22, 82.27it/s] 33%|███▎ | 882/2708 [00:11<00:22, 81.32it/s] 33%|███▎ | 891/2708 [00:11<00:22, 81.65it/s] 33%|███▎ | 900/2708 [00:11<00:22, 81.88it/s] 34%|███▎ | 909/2708 [00:11<00:21, 82.03it/s] 34%|███▍ | 918/2708 [00:11<00:21, 81.85it/s] 34%|███▍ | 927/2708 [00:11<00:21, 81.99it/s] 35%|███▍ | 936/2708 [00:11<00:21, 81.75it/s] 35%|███▍ | 945/2708 [00:12<00:21, 81.73it/s] 35%|███▌ | 954/2708 [00:12<00:21, 81.79it/s] 36%|███▌ | 963/2708 [00:12<00:21, 82.03it/s] 36%|███▌ | 972/2708 [00:12<00:21, 82.17it/s] 36%|███▌ | 981/2708 [00:12<00:21, 82.09it/s] 37%|███▋ | 990/2708 [00:12<00:21, 81.57it/s] 37%|███▋ | 999/2708 [00:12<00:21, 81.36it/s] 37%|███▋ | 1008/2708 [00:12<00:20, 81.71it/s] 38%|███▊ | 1017/2708 [00:12<00:20, 81.98it/s] 38%|███▊ | 1026/2708 [00:13<00:20, 82.18it/s] 38%|███▊ | 1035/2708 [00:13<00:20, 82.29it/s] 39%|███▊ | 1044/2708 [00:13<00:20, 81.75it/s] 39%|███▉ | 1053/2708 [00:13<00:20, 81.97it/s] 39%|███▉ | 1062/2708 [00:13<00:20, 82.15it/s] 40%|███▉ | 1071/2708 [00:13<00:19, 81.98it/s] 40%|███▉ | 1080/2708 [00:13<00:19, 82.21it/s] 40%|████ | 1089/2708 [00:13<00:19, 82.28it/s] 41%|████ | 1098/2708 [00:13<00:19, 81.92it/s] 41%|████ | 1107/2708 [00:14<00:19, 81.49it/s] 41%|████ | 1116/2708 [00:14<00:19, 81.83it/s] 42%|████▏ | 1125/2708 [00:14<00:19, 82.21it/s] 42%|████▏ | 1134/2708 [00:14<00:19, 82.38it/s] 42%|████▏ | 1143/2708 [00:14<00:18, 82.45it/s] 43%|████▎ | 1152/2708 [00:14<00:19, 81.83it/s] 43%|████▎ | 1161/2708 [00:14<00:18, 81.98it/s] 43%|████▎ | 1170/2708 [00:14<00:18, 81.98it/s] 44%|████▎ | 1179/2708 [00:14<00:18, 82.12it/s] 44%|████▍ | 1188/2708 [00:14<00:18, 82.24it/s] 44%|████▍ | 1197/2708 [00:15<00:18, 82.35it/s] 45%|████▍ | 1206/2708 [00:15<00:19, 78.96it/s] 45%|████▍ | 1215/2708 [00:15<00:18, 79.78it/s] 45%|████▌ | 1224/2708 [00:15<00:18, 80.61it/s] 46%|████▌ | 1233/2708 [00:15<00:18, 80.99it/s] 46%|████▌ | 1242/2708 [00:15<00:18, 81.23it/s] 46%|████▌ | 1251/2708 [00:15<00:17, 81.59it/s] 47%|████▋ | 1260/2708 [00:15<00:17, 81.24it/s] 47%|████▋ | 1269/2708 [00:15<00:17, 81.62it/s] 47%|████▋ | 1278/2708 [00:16<00:17, 81.85it/s] 48%|████▊ | 1287/2708 [00:16<00:17, 81.90it/s] 48%|████▊ | 1296/2708 [00:16<00:17, 81.85it/s] 48%|████▊ | 1305/2708 [00:16<00:17, 81.88it/s] 49%|████▊ | 1314/2708 [00:16<00:17, 81.42it/s] 49%|████▉ | 1323/2708 [00:16<00:17, 81.26it/s] 49%|████▉ | 1332/2708 [00:16<00:16, 81.50it/s] 50%|████▉ | 1341/2708 [00:16<00:16, 81.62it/s] 50%|████▉ | 1350/2708 [00:16<00:16, 81.78it/s] 50%|█████ | 1359/2708 [00:17<00:16, 81.64it/s] 51%|█████ | 1368/2708 [00:17<00:16, 81.69it/s] 51%|█████ | 1377/2708 [00:17<00:16, 81.91it/s] 51%|█████ | 1386/2708 [00:17<00:16, 81.90it/s] 52%|█████▏ | 1395/2708 [00:17<00:16, 81.84it/s] 52%|█████▏ | 1404/2708 [00:17<00:15, 82.08it/s] 52%|█████▏ | 1413/2708 [00:17<00:15, 81.78it/s] 53%|█████▎ | 1422/2708 [00:17<00:15, 81.78it/s] 53%|█████▎ | 1431/2708 [00:17<00:15, 81.37it/s] 53%|█████▎ | 1440/2708 [00:18<00:15, 81.53it/s] 54%|█████▎ | 1449/2708 [00:18<00:15, 81.80it/s] 54%|█████▍ | 1458/2708 [00:18<00:15, 81.84it/s] 54%|█████▍ | 1467/2708 [00:18<00:15, 81.76it/s] 55%|█████▍ | 1476/2708 [00:18<00:15, 81.64it/s] 55%|█████▍ | 1485/2708 [00:18<00:14, 81.64it/s] 55%|█████▌ | 1494/2708 [00:18<00:14, 81.70it/s] 56%|█████▌ | 1503/2708 [00:18<00:14, 81.79it/s] 56%|█████▌ | 1512/2708 [00:18<00:14, 81.96it/s] 56%|█████▌ | 1521/2708 [00:19<00:14, 81.66it/s] 56%|█████▋ | 1530/2708 [00:19<00:14, 81.57it/s] 57%|█████▋ | 1539/2708 [00:19<00:14, 80.99it/s] 57%|█████▋ | 1548/2708 [00:19<00:14, 81.45it/s] 57%|█████▋ | 1557/2708 [00:19<00:14, 81.79it/s] 58%|█████▊ | 1566/2708 [00:19<00:13, 82.02it/s] 58%|█████▊ | 1575/2708 [00:19<00:13, 81.74it/s] 58%|█████▊ | 1584/2708 [00:19<00:13, 81.94it/s] 59%|█████▉ | 1593/2708 [00:19<00:13, 81.93it/s] 59%|█████▉ | 1602/2708 [00:20<00:13, 82.27it/s] 59%|█████▉ | 1611/2708 [00:20<00:13, 82.26it/s] 60%|█████▉ | 1620/2708 [00:20<00:13, 82.10it/s] 60%|██████ | 1629/2708 [00:20<00:13, 81.82it/s] 60%|██████ | 1638/2708 [00:20<00:13, 81.94it/s] 61%|██████ | 1647/2708 [00:20<00:12, 81.95it/s] 61%|██████ | 1656/2708 [00:20<00:12, 81.53it/s] 61%|██████▏ | 1665/2708 [00:20<00:12, 81.80it/s] 62%|██████▏ | 1674/2708 [00:20<00:12, 81.75it/s] 62%|██████▏ | 1683/2708 [00:21<00:12, 81.26it/s] 62%|██████▏ | 1692/2708 [00:21<00:12, 81.48it/s] 63%|██████▎ | 1701/2708 [00:21<00:12, 81.62it/s] 63%|██████▎ | 1710/2708 [00:21<00:12, 81.94it/s] 63%|██████▎ | 1719/2708 [00:21<00:12, 81.95it/s] 64%|██████▍ | 1728/2708 [00:21<00:11, 82.09it/s] 64%|██████▍ | 1737/2708 [00:21<00:11, 81.55it/s] 64%|██████▍ | 1746/2708 [00:21<00:11, 81.82it/s] 65%|██████▍ | 1755/2708 [00:21<00:11, 82.07it/s] 65%|██████▌ | 1764/2708 [00:22<00:11, 81.98it/s] 65%|██████▌ | 1773/2708 [00:22<00:11, 81.92it/s] 66%|██████▌ | 1782/2708 [00:22<00:11, 82.11it/s] 66%|██████▌ | 1791/2708 [00:22<00:11, 81.53it/s] 66%|██████▋ | 1800/2708 [00:22<00:11, 81.44it/s] 67%|██████▋ | 1809/2708 [00:22<00:11, 81.56it/s] 67%|██████▋ | 1818/2708 [00:22<00:10, 81.83it/s] 67%|██████▋ | 1827/2708 [00:22<00:10, 82.05it/s] 68%|██████▊ | 1836/2708 [00:22<00:10, 81.83it/s] 68%|██████▊ | 1845/2708 [00:23<00:10, 78.67it/s] 68%|██████▊ | 1854/2708 [00:23<00:10, 79.93it/s] 69%|██████▉ | 1863/2708 [00:23<00:10, 80.98it/s] 69%|██████▉ | 1872/2708 [00:23<00:10, 80.68it/s] 69%|██████▉ | 1881/2708 [00:23<00:10, 81.09it/s] 70%|██████▉ | 1890/2708 [00:23<00:10, 81.49it/s] 70%|███████ | 1899/2708 [00:23<00:09, 81.14it/s] 70%|███████ | 1908/2708 [00:23<00:09, 81.58it/s] 71%|███████ | 1917/2708 [00:23<00:09, 81.66it/s] 71%|███████ | 1926/2708 [00:24<00:09, 81.72it/s] 71%|███████▏ | 1935/2708 [00:24<00:09, 81.78it/s] 72%|███████▏ | 1944/2708 [00:24<00:09, 82.03it/s] 72%|███████▏ | 1953/2708 [00:24<00:09, 81.21it/s] 72%|███████▏ | 1962/2708 [00:24<00:09, 81.64it/s] 73%|███████▎ | 1971/2708 [00:24<00:08, 82.10it/s] 73%|███████▎ | 1980/2708 [00:24<00:08, 81.62it/s] 73%|███████▎ | 1989/2708 [00:24<00:08, 81.95it/s] 74%|███████▍ | 1998/2708 [00:24<00:08, 82.10it/s] 74%|███████▍ | 2007/2708 [00:25<00:08, 81.51it/s] 74%|███████▍ | 2016/2708 [00:25<00:08, 81.82it/s] 75%|███████▍ | 2025/2708 [00:25<00:08, 82.04it/s] 75%|███████▌ | 2034/2708 [00:25<00:08, 82.24it/s] 75%|███████▌ | 2043/2708 [00:25<00:08, 82.33it/s] 76%|███████▌ | 2052/2708 [00:25<00:07, 82.23it/s] 76%|███████▌ | 2061/2708 [00:25<00:07, 81.86it/s] 76%|███████▋ | 2070/2708 [00:25<00:07, 82.06it/s] 77%|███████▋ | 2079/2708 [00:25<00:07, 82.19it/s] 77%|███████▋ | 2088/2708 [00:26<00:07, 81.83it/s] 77%|███████▋ | 2097/2708 [00:26<00:07, 82.05it/s] 78%|███████▊ | 2106/2708 [00:26<00:07, 82.03it/s] 78%|███████▊ | 2115/2708 [00:26<00:07, 81.73it/s] 78%|███████▊ | 2124/2708 [00:26<00:07, 81.96it/s] 79%|███████▉ | 2133/2708 [00:26<00:07, 81.95it/s] 79%|███████▉ | 2142/2708 [00:26<00:06, 82.15it/s] 79%|███████▉ | 2151/2708 [00:26<00:06, 82.24it/s] 80%|███████▉ | 2160/2708 [00:26<00:06, 82.38it/s] 80%|████████ | 2169/2708 [00:27<00:06, 81.95it/s] 80%|████████ | 2178/2708 [00:27<00:06, 82.13it/s] 81%|████████ | 2187/2708 [00:27<00:06, 82.29it/s] 81%|████████ | 2196/2708 [00:27<00:06, 82.58it/s] 81%|████████▏ | 2205/2708 [00:27<00:06, 81.67it/s] 82%|████████▏ | 2214/2708 [00:27<00:06, 81.91it/s] 82%|████████▏ | 2223/2708 [00:27<00:05, 81.68it/s] 82%|████████▏ | 2232/2708 [00:27<00:05, 81.91it/s] 83%|████████▎ | 2241/2708 [00:27<00:05, 82.16it/s] 83%|████████▎ | 2250/2708 [00:27<00:05, 82.27it/s] 83%|████████▎ | 2259/2708 [00:28<00:05, 82.34it/s] 84%|████████▍ | 2268/2708 [00:28<00:05, 82.25it/s] 84%|████████▍ | 2277/2708 [00:28<00:05, 81.90it/s] 84%|████████▍ | 2286/2708 [00:28<00:05, 82.21it/s] 85%|████████▍ | 2295/2708 [00:28<00:05, 82.34it/s] 85%|████████▌ | 2304/2708 [00:28<00:04, 82.18it/s] 85%|████████▌ | 2313/2708 [00:28<00:04, 81.67it/s] 86%|████████▌ | 2322/2708 [00:28<00:04, 81.96it/s] 86%|████████▌ | 2331/2708 [00:28<00:04, 81.67it/s] 86%|████████▋ | 2340/2708 [00:29<00:04, 81.94it/s] 87%|████████▋ | 2349/2708 [00:29<00:04, 81.95it/s] 87%|████████▋ | 2358/2708 [00:29<00:04, 82.37it/s] 87%|████████▋ | 2367/2708 [00:29<00:04, 82.40it/s] 88%|████████▊ | 2376/2708 [00:29<00:04, 82.20it/s] 88%|████████▊ | 2385/2708 [00:29<00:03, 81.63it/s] 88%|████████▊ | 2394/2708 [00:29<00:03, 81.69it/s] 89%|████████▊ | 2403/2708 [00:29<00:03, 81.92it/s] 89%|████████▉ | 2412/2708 [00:29<00:03, 82.14it/s] 89%|████████▉ | 2421/2708 [00:30<00:03, 81.56it/s] 90%|████████▉ | 2430/2708 [00:30<00:03, 81.85it/s] 90%|█████████ | 2439/2708 [00:30<00:03, 81.45it/s] 90%|█████████ | 2448/2708 [00:30<00:03, 81.74it/s] 91%|█████████ | 2457/2708 [00:30<00:03, 82.02it/s] 91%|█████████ | 2466/2708 [00:30<00:02, 82.19it/s] 91%|█████████▏| 2475/2708 [00:30<00:02, 82.53it/s] 92%|█████████▏| 2484/2708 [00:30<00:02, 82.55it/s] 92%|█████████▏| 2493/2708 [00:30<00:02, 79.06it/s] 92%|█████████▏| 2502/2708 [00:31<00:02, 80.34it/s] 93%|█████████▎| 2511/2708 [00:31<00:02, 81.24it/s] 93%|█████████▎| 2520/2708 [00:31<00:02, 81.56it/s] 93%|█████████▎| 2529/2708 [00:31<00:02, 81.59it/s] 94%|█████████▎| 2538/2708 [00:31<00:02, 81.74it/s] 94%|█████████▍| 2547/2708 [00:31<00:01, 81.52it/s] 94%|█████████▍| 2556/2708 [00:31<00:01, 81.77it/s] 95%|█████████▍| 2565/2708 [00:31<00:01, 81.85it/s] 95%|█████████▌| 2574/2708 [00:31<00:01, 82.03it/s] 95%|█████████▌| 2583/2708 [00:32<00:01, 82.21it/s] 96%|█████████▌| 2592/2708 [00:32<00:01, 82.30it/s] 96%|█████████▌| 2601/2708 [00:32<00:01, 81.75it/s] 96%|█████████▋| 2610/2708 [00:32<00:01, 82.01it/s] 97%|█████████▋| 2619/2708 [00:32<00:01, 82.22it/s] 97%|█████████▋| 2628/2708 [00:32<00:00, 82.29it/s] 97%|█████████▋| 2637/2708 [00:32<00:00, 82.07it/s] 98%|█████████▊| 2646/2708 [00:32<00:00, 82.00it/s] 98%|█████████▊| 2655/2708 [00:32<00:00, 81.72it/s] 98%|█████████▊| 2664/2708 [00:33<00:00, 81.56it/s] 99%|█████████▊| 2673/2708 [00:33<00:00, 81.87it/s] 99%|█████████▉| 2682/2708 [00:33<00:00, 81.89it/s] 99%|█████████▉| 2691/2708 [00:33<00:00, 82.25it/s] 100%|█████████▉| 2700/2708 [00:33<00:00, 82.39it/s] 100%|██████████| 2708/2708 [00:33<00:00, 80.64it/s]
INFO:TestAccuracy:Batch size is 4, F1: 80.15325, Exact Match:72.95975
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
2024-10-29 10:23:39.494164276 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-10-29 10:23:39.494180436 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 4, QPS: 375, Avg Latency:10.66, Tail Latency:12.69
2024-10-29 10:23:41.914490180 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-10-29 10:23:41.914506778 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 8, QPS: 448, Avg Latency:17.82, Tail Latency:19.7
2024-10-29 10:23:45.306684295 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-10-29 10:23:45.306699721 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 16, QPS: 486, Avg Latency:32.92, Tail Latency:34.87
2024-10-29 10:23:50.633233948 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-10-29 10:23:50.633250336 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 32, QPS: 514, Avg Latency:62.17, Tail Latency:63.92
2024-10-29 10:23:59.768382259 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-10-29 10:23:59.768398238 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 64, QPS: 537, Avg Latency:119.1, Tail Latency:121.3
2024-10-29 10:24:16.353528795 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-10-29 10:24:16.353545488 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 128, QPS: 514, Avg Latency:248.69, Tail Latency:277.44
INFO:PerfEngine:Testing Finish. Report is saved in path: [ general_perf/reports/DCU/bert-onnxruntime-fp16/result-fp16.json ]
INFO:PerfEngine:PDF Version is saved in path: [ general_perf/reports/DCU/bert-onnxruntime-fp16/BERT-ONNXRUNTIME-FP16-TO-FP16.JSON.pdf ]
Writing predictions to: /home/workspace/ByteMLPerf/byte_infer_perf/general_perf/reports/DCU/predictions.json
ByteMLPerf/byte_infer_perf/general_perf/log/bert-onnxruntime-fp32.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
INFO:PerfEngine:******************************************* Start to test model: bert-onnxruntime-fp32. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_squad
INFO:SQUAD:Initial...
INFO:SQUAD:Preprocessing...
INFO:SQUAD:Rebatching batch size to: 10833 ...
0%| | 0/1 [00:00<?, ?it/s] 100%|██████████| 1/1 [00:00<00:00, 98.69it/s]
INFO:PerfEngine:Start to compile the model...
2024-10-29 10:25:05.286401290 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-10-29 10:25:05.286419867 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:SQUAD:Rebatching batch size to: 4 ...
0%| | 0/2708 [00:00<?, ?it/s] 100%|██████████| 2708/2708 [00:00<00:00, 135118.25it/s]
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/2708 [00:00<?, ?it/s] 0%| | 1/2708 [00:00<20:41, 2.18it/s] 0%| | 6/2708 [00:00<03:20, 13.45it/s] 0%| | 11/2708 [00:00<02:00, 22.46it/s] 1%| | 16/2708 [00:00<01:31, 29.27it/s] 1%| | 21/2708 [00:00<01:17, 34.54it/s] 1%| | 26/2708 [00:00<01:09, 38.46it/s] 1%| | 31/2708 [00:01<01:04, 41.31it/s] 1%|▏ | 36/2708 [00:01<01:01, 43.38it/s] 2%|▏ | 41/2708 [00:01<00:59, 44.78it/s] 2%|▏ | 46/2708 [00:01<00:58, 45.80it/s] 2%|▏ | 51/2708 [00:01<00:57, 46.37it/s] 2%|▏ | 56/2708 [00:01<00:56, 46.81it/s] 2%|▏ | 61/2708 [00:01<00:56, 47.12it/s] 2%|▏ | 66/2708 [00:01<00:55, 47.36it/s] 3%|▎ | 71/2708 [00:01<00:55, 47.33it/s] 3%|▎ | 76/2708 [00:02<00:55, 47.29it/s] 3%|▎ | 81/2708 [00:02<00:55, 47.49it/s] 3%|▎ | 86/2708 [00:02<00:54, 47.76it/s] 3%|▎ | 91/2708 [00:02<00:54, 47.91it/s] 4%|▎ | 96/2708 [00:02<00:54, 48.06it/s] 4%|▎ | 101/2708 [00:02<00:54, 48.05it/s] 4%|▍ | 106/2708 [00:02<00:54, 48.08it/s] 4%|▍ | 111/2708 [00:02<00:54, 48.02it/s] 4%|▍ | 116/2708 [00:02<00:53, 48.02it/s] 4%|▍ | 121/2708 [00:02<00:53, 48.03it/s] 5%|▍ | 126/2708 [00:03<00:53, 47.89it/s] 5%|▍ | 131/2708 [00:03<00:53, 47.95it/s] 5%|▌ | 136/2708 [00:03<00:53, 48.05it/s] 5%|▌ | 141/2708 [00:03<00:53, 47.76it/s] 5%|▌ | 146/2708 [00:03<00:53, 47.85it/s] 6%|▌ | 151/2708 [00:03<00:53, 48.00it/s] 6%|▌ | 156/2708 [00:03<00:53, 47.98it/s] 6%|▌ | 161/2708 [00:03<00:53, 47.98it/s] 6%|▌ | 166/2708 [00:03<00:52, 48.06it/s] 6%|▋ | 171/2708 [00:03<00:52, 48.06it/s] 6%|▋ | 176/2708 [00:04<00:52, 47.94it/s] 7%|▋ | 181/2708 [00:04<00:52, 47.99it/s] 7%|▋ | 186/2708 [00:04<00:52, 48.02it/s] 7%|▋ | 191/2708 [00:04<00:52, 48.05it/s] 7%|▋ | 196/2708 [00:04<00:52, 48.05it/s] 7%|▋ | 201/2708 [00:04<00:52, 48.06it/s] 8%|▊ | 206/2708 [00:04<00:52, 47.80it/s] 8%|▊ | 211/2708 [00:04<00:52, 47.84it/s] 8%|▊ | 216/2708 [00:04<00:52, 47.90it/s] 8%|▊ | 221/2708 [00:05<00:51, 47.92it/s] 8%|▊ | 226/2708 [00:05<00:51, 48.05it/s] 9%|▊ | 231/2708 [00:05<00:51, 47.85it/s] 9%|▊ | 236/2708 [00:05<00:51, 47.84it/s] 9%|▉ | 241/2708 [00:05<00:51, 47.91it/s] 9%|▉ | 246/2708 [00:05<00:51, 47.92it/s] 9%|▉ | 251/2708 [00:05<00:51, 48.00it/s] 9%|▉ | 256/2708 [00:05<00:51, 47.91it/s] 10%|▉ | 261/2708 [00:05<00:51, 47.95it/s] 10%|▉ | 266/2708 [00:05<00:50, 47.96it/s] 10%|█ | 271/2708 [00:06<00:50, 47.82it/s] 10%|█ | 276/2708 [00:06<00:50, 47.86it/s] 10%|█ | 281/2708 [00:06<00:50, 47.92it/s] 11%|█ | 286/2708 [00:06<00:50, 47.93it/s] 11%|█ | 291/2708 [00:06<00:50, 48.00it/s] 11%|█ | 296/2708 [00:06<00:50, 48.12it/s] 11%|█ | 301/2708 [00:06<00:49, 48.22it/s] 11%|█▏ | 306/2708 [00:06<00:49, 48.20it/s] 11%|█▏ | 311/2708 [00:06<00:49, 48.18it/s] 12%|█▏ | 316/2708 [00:07<00:49, 48.17it/s] 12%|█▏ | 321/2708 [00:07<00:49, 48.11it/s] 12%|█▏ | 326/2708 [00:07<00:49, 48.11it/s] 12%|█▏ | 331/2708 [00:07<00:49, 48.14it/s] 12%|█▏ | 336/2708 [00:07<00:49, 47.89it/s] 13%|█▎ | 341/2708 [00:07<00:49, 47.87it/s] 13%|█▎ | 346/2708 [00:07<00:49, 47.94it/s] 13%|█▎ | 351/2708 [00:07<00:49, 47.98it/s] 13%|█▎ | 356/2708 [00:07<00:49, 48.00it/s] 13%|█▎ | 361/2708 [00:07<00:48, 48.05it/s] 14%|█▎ | 366/2708 [00:08<00:48, 48.03it/s] 14%|█▎ | 371/2708 [00:08<00:48, 48.06it/s] 14%|█▍ | 376/2708 [00:08<00:48, 48.14it/s] 14%|█▍ | 381/2708 [00:08<00:48, 48.18it/s] 14%|█▍ | 386/2708 [00:08<00:48, 48.18it/s] 14%|█▍ | 391/2708 [00:08<00:48, 48.10it/s] 15%|█▍ | 396/2708 [00:08<00:47, 48.19it/s] 15%|█▍ | 401/2708 [00:08<00:48, 47.95it/s] 15%|█▍ | 406/2708 [00:08<00:47, 48.07it/s] 15%|█▌ | 411/2708 [00:08<00:47, 48.05it/s] 15%|█▌ | 416/2708 [00:09<00:47, 48.10it/s] 16%|█▌ | 421/2708 [00:09<00:47, 48.13it/s] 16%|█▌ | 426/2708 [00:09<00:47, 48.14it/s] 16%|█▌ | 431/2708 [00:09<00:47, 48.16it/s] 16%|█▌ | 436/2708 [00:09<00:47, 48.21it/s] 16%|█▋ | 441/2708 [00:09<00:46, 48.24it/s] 16%|█▋ | 446/2708 [00:09<00:47, 48.06it/s] 17%|█▋ | 451/2708 [00:09<00:46, 48.07it/s] 17%|█▋ | 456/2708 [00:09<00:46, 48.12it/s] 17%|█▋ | 461/2708 [00:10<00:46, 48.08it/s] 17%|█▋ | 466/2708 [00:10<00:46, 47.82it/s] 17%|█▋ | 471/2708 [00:10<00:46, 47.94it/s] 18%|█▊ | 476/2708 [00:10<00:46, 47.96it/s] 18%|█▊ | 481/2708 [00:10<00:46, 48.14it/s] 18%|█▊ | 486/2708 [00:10<00:46, 48.18it/s] 18%|█▊ | 491/2708 [00:10<00:46, 48.20it/s] 18%|█▊ | 496/2708 [00:10<00:45, 48.19it/s] 19%|█▊ | 501/2708 [00:10<00:45, 48.07it/s] 19%|█▊ | 506/2708 [00:10<00:45, 48.00it/s] 19%|█▉ | 511/2708 [00:11<00:45, 48.04it/s] 19%|█▉ | 516/2708 [00:11<00:45, 48.16it/s] 19%|█▉ | 521/2708 [00:11<00:45, 48.16it/s] 19%|█▉ | 526/2708 [00:11<00:45, 48.22it/s] 20%|█▉ | 531/2708 [00:11<00:45, 48.01it/s] 20%|█▉ | 536/2708 [00:11<00:45, 48.06it/s] 20%|█▉ | 541/2708 [00:11<00:45, 47.97it/s] 20%|██ | 546/2708 [00:11<00:44, 48.09it/s] 20%|██ | 551/2708 [00:11<00:44, 48.07it/s] 21%|██ | 556/2708 [00:12<00:46, 46.60it/s] 21%|██ | 561/2708 [00:12<00:45, 47.07it/s] 21%|██ | 566/2708 [00:12<00:45, 47.42it/s] 21%|██ | 571/2708 [00:12<00:44, 47.60it/s] 21%|██▏ | 576/2708 [00:12<00:44, 47.74it/s] 21%|██▏ | 581/2708 [00:12<00:44, 47.84it/s] 22%|██▏ | 586/2708 [00:12<00:44, 47.92it/s] 22%|██▏ | 591/2708 [00:12<00:44, 47.82it/s] 22%|██▏ | 596/2708 [00:12<00:44, 47.96it/s] 22%|██▏ | 601/2708 [00:12<00:43, 48.05it/s] 22%|██▏ | 606/2708 [00:13<00:43, 47.82it/s] 23%|██▎ | 611/2708 [00:13<00:43, 47.95it/s] 23%|██▎ | 616/2708 [00:13<00:43, 48.02it/s] 23%|██▎ | 621/2708 [00:13<00:43, 48.01it/s] 23%|██▎ | 626/2708 [00:13<00:43, 48.07it/s] 23%|██▎ | 631/2708 [00:13<00:43, 48.14it/s] 23%|██▎ | 636/2708 [00:13<00:43, 48.04it/s] 24%|██▎ | 641/2708 [00:13<00:43, 48.06it/s] 24%|██▍ | 646/2708 [00:13<00:42, 48.04it/s] 24%|██▍ | 651/2708 [00:14<00:42, 47.91it/s] 24%|██▍ | 656/2708 [00:14<00:43, 47.65it/s] 24%|██▍ | 661/2708 [00:14<00:42, 47.72it/s] 25%|██▍ | 666/2708 [00:14<00:42, 47.82it/s] 25%|██▍ | 671/2708 [00:14<00:42, 47.88it/s] 25%|██▍ | 676/2708 [00:14<00:42, 47.91it/s] 25%|██▌ | 681/2708 [00:14<00:42, 48.00it/s] 25%|██▌ | 686/2708 [00:14<00:42, 48.04it/s] 26%|██▌ | 691/2708 [00:14<00:42, 48.02it/s] 26%|██▌ | 696/2708 [00:14<00:41, 48.08it/s] 26%|██▌ | 701/2708 [00:15<00:41, 48.17it/s] 26%|██▌ | 706/2708 [00:15<00:41, 48.17it/s] 26%|██▋ | 711/2708 [00:15<00:41, 48.15it/s] 26%|██▋ | 716/2708 [00:15<00:41, 47.97it/s] 27%|██▋ | 721/2708 [00:15<00:41, 47.65it/s] 27%|██▋ | 726/2708 [00:15<00:41, 47.87it/s] 27%|██▋ | 731/2708 [00:15<00:41, 47.92it/s] 27%|██▋ | 736/2708 [00:15<00:41, 47.96it/s] 27%|██▋ | 741/2708 [00:15<00:41, 47.97it/s] 28%|██▊ | 746/2708 [00:15<00:40, 48.11it/s] 28%|██▊ | 751/2708 [00:16<00:40, 48.17it/s] 28%|██▊ | 756/2708 [00:16<00:40, 48.12it/s] 28%|██▊ | 761/2708 [00:16<00:40, 48.12it/s] 28%|██▊ | 766/2708 [00:16<00:40, 48.08it/s] 28%|██▊ | 771/2708 [00:16<00:40, 47.93it/s] 29%|██▊ | 776/2708 [00:16<00:40, 47.95it/s] 29%|██▉ | 781/2708 [00:16<00:40, 47.98it/s] 29%|██▉ | 786/2708 [00:16<00:40, 47.79it/s] 29%|██▉ | 791/2708 [00:16<00:40, 47.88it/s] 29%|██▉ | 796/2708 [00:17<00:40, 47.79it/s] 30%|██▉ | 801/2708 [00:17<00:39, 47.86it/s] 30%|██▉ | 806/2708 [00:17<00:39, 47.91it/s] 30%|██▉ | 811/2708 [00:17<00:39, 48.05it/s] 30%|███ | 816/2708 [00:17<00:39, 47.99it/s] 30%|███ | 821/2708 [00:17<00:39, 47.86it/s] 31%|███ | 826/2708 [00:17<00:39, 47.95it/s] 31%|███ | 831/2708 [00:17<00:39, 47.93it/s] 31%|███ | 836/2708 [00:17<00:38, 48.04it/s] 31%|███ | 841/2708 [00:17<00:38, 48.05it/s] 31%|███ | 846/2708 [00:18<00:38, 48.00it/s] 31%|███▏ | 851/2708 [00:18<00:38, 47.80it/s] 32%|███▏ | 856/2708 [00:18<00:38, 47.84it/s] 32%|███▏ | 861/2708 [00:18<00:38, 47.96it/s] 32%|███▏ | 866/2708 [00:18<00:38, 48.01it/s] 32%|███▏ | 871/2708 [00:18<00:38, 48.03it/s] 32%|███▏ | 876/2708 [00:18<00:38, 47.88it/s] 33%|███▎ | 881/2708 [00:18<00:38, 47.83it/s] 33%|███▎ | 886/2708 [00:18<00:38, 47.87it/s] 33%|███▎ | 891/2708 [00:19<00:37, 47.90it/s] 33%|███▎ | 896/2708 [00:19<00:37, 47.99it/s] 33%|███▎ | 901/2708 [00:19<00:37, 47.93it/s] 33%|███▎ | 906/2708 [00:19<00:37, 47.96it/s] 34%|███▎ | 911/2708 [00:19<00:37, 47.94it/s] 34%|███▍ | 916/2708 [00:19<00:37, 47.67it/s] 34%|███▍ | 921/2708 [00:19<00:37, 47.88it/s] 34%|███▍ | 926/2708 [00:19<00:37, 47.91it/s] 34%|███▍ | 931/2708 [00:19<00:37, 47.84it/s] 35%|███▍ | 936/2708 [00:19<00:37, 47.80it/s] 35%|███▍ | 941/2708 [00:20<00:36, 47.87it/s] 35%|███▍ | 946/2708 [00:20<00:36, 47.81it/s]
\ No newline at end of file
ByteMLPerf/byte_infer_perf/general_perf/log/bert-torch-fp16.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
/usr/local/lib/python3.10/site-packages/tensorflow/python/keras/engine/training_arrays_v1.py:37: UserWarning: A NumPy version >=1.23.5 and <2.3.0 is required for this version of SciPy (detected version 1.23.0)
from scipy.sparse import issparse # pylint: disable=g-import-not-at-top
INFO:PerfEngine:******************************************* Start to test model: bert-torch-fp16. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_squad
INFO:SQUAD:Initial...
INFO:SQUAD:Preprocessing...
INFO:SQUAD:Rebatching batch size to: 10833 ...
0%| | 0/1 [00:00<?, ?it/s] 100%|██████████| 1/1 [00:00<00:00, 100.84it/s]
INFO:PerfEngine:Start to compile the model...
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:SQUAD:Rebatching batch size to: 1 ...
0%| | 0/10833 [00:00<?, ?it/s] 75%|███████▌ | 8136/10833 [00:00<00:00, 28782.63it/s] 100%|██████████| 10833/10833 [00:00<00:00, 37516.77it/s]
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/10833 [00:00<?, ?it/s] 0%| | 1/10833 [00:04<12:29:12, 4.15s/it] 0%| | 3/10833 [00:04<3:21:21, 1.12s/it] 0%| | 16/10833 [00:04<26:45, 6.74it/s] 0%| | 29/10833 [00:04<12:39, 14.22it/s] 0%| | 42/10833 [00:04<07:40, 23.44it/s] 1%| | 55/10833 [00:04<05:14, 34.29it/s] 1%| | 68/10833 [00:04<03:51, 46.43it/s] 1%| | 81/10833 [00:04<03:01, 59.13it/s] 1%| | 94/10833 [00:04<02:29, 71.74it/s] 1%| | 107/10833 [00:05<02:08, 83.41it/s] 1%| | 120/10833 [00:05<01:55, 93.13it/s] 1%| | 133/10833 [00:05<01:45, 101.77it/s] 1%|▏ | 146/10833 [00:05<01:38, 108.68it/s] 1%|▏ | 159/10833 [00:05<01:33, 114.06it/s] 2%|▏ | 172/10833 [00:05<01:30, 118.06it/s] 2%|▏ | 185/10833 [00:05<01:27, 121.04it/s] 2%|▏ | 198/10833 [00:05<01:26, 123.18it/s] 2%|▏ | 211/10833 [00:05<01:25, 124.76it/s] 2%|▏ | 224/10833 [00:05<01:24, 125.78it/s] 2%|▏ | 237/10833 [00:06<01:23, 126.60it/s] 2%|▏ | 250/10833 [00:06<01:23, 126.65it/s] 2%|▏ | 263/10833 [00:06<01:23, 127.12it/s] 3%|▎ | 276/10833 [00:06<01:22, 127.48it/s] 3%|▎ | 289/10833 [00:06<01:23, 126.75it/s] 3%|▎ | 302/10833 [00:06<01:22, 127.33it/s] 3%|▎ | 315/10833 [00:06<01:22, 127.68it/s] 3%|▎ | 328/10833 [00:06<01:22, 127.95it/s] 3%|▎ | 341/10833 [00:06<01:21, 128.08it/s] 3%|▎ | 354/10833 [00:06<01:21, 128.11it/s] 3%|▎ | 367/10833 [00:07<01:21, 128.18it/s] 4%|▎ | 380/10833 [00:07<01:21, 128.19it/s] 4%|▎ | 393/10833 [00:07<01:21, 128.16it/s] 4%|▎ | 406/10833 [00:07<01:21, 128.13it/s] 4%|▍ | 419/10833 [00:07<01:21, 127.82it/s] 4%|▍ | 432/10833 [00:07<01:21, 128.04it/s] 4%|▍ | 445/10833 [00:07<01:21, 128.16it/s] 4%|▍ | 458/10833 [00:07<01:21, 128.06it/s] 4%|▍ | 471/10833 [00:07<01:21, 127.31it/s] 4%|▍ | 484/10833 [00:08<01:21, 127.71it/s] 5%|▍ | 497/10833 [00:08<01:20, 127.89it/s] 5%|▍ | 510/10833 [00:08<01:20, 128.02it/s] 5%|▍ | 523/10833 [00:08<01:20, 127.97it/s] 5%|▍ | 536/10833 [00:08<01:20, 128.12it/s] 5%|▌ | 549/10833 [00:08<01:20, 128.17it/s] 5%|▌ | 562/10833 [00:08<01:20, 128.22it/s] 5%|▌ | 575/10833 [00:08<01:19, 128.31it/s] 5%|▌ | 588/10833 [00:08<01:19, 128.28it/s] 6%|▌ | 601/10833 [00:08<01:19, 128.33it/s] 6%|▌ | 614/10833 [00:09<01:19, 128.26it/s] 6%|▌ | 627/10833 [00:09<01:19, 128.15it/s] 6%|▌ | 640/10833 [00:09<01:20, 127.21it/s] 6%|▌ | 653/10833 [00:09<01:19, 127.40it/s] 6%|▌ | 666/10833 [00:09<01:19, 127.49it/s] 6%|▋ | 679/10833 [00:09<01:19, 127.73it/s] 6%|▋ | 692/10833 [00:09<01:19, 127.84it/s] 7%|▋ | 705/10833 [00:09<01:19, 127.98it/s] 7%|▋ | 718/10833 [00:09<01:18, 128.10it/s] 7%|▋ | 731/10833 [00:09<01:18, 128.13it/s] 7%|▋ | 744/10833 [00:10<01:18, 128.18it/s] 7%|▋ | 757/10833 [00:10<01:18, 128.19it/s] 7%|▋ | 770/10833 [00:10<01:18, 127.67it/s] 7%|▋ | 783/10833 [00:10<01:18, 127.84it/s] 7%|▋ | 796/10833 [00:10<01:18, 127.97it/s] 7%|▋ | 809/10833 [00:10<01:18, 127.23it/s] 8%|▊ | 822/10833 [00:10<01:18, 127.55it/s] 8%|▊ | 835/10833 [00:10<01:18, 127.69it/s] 8%|▊ | 848/10833 [00:10<01:18, 127.85it/s] 8%|▊ | 861/10833 [00:10<01:17, 127.99it/s] 8%|▊ | 874/10833 [00:11<01:17, 128.06it/s] 8%|▊ | 887/10833 [00:11<01:17, 128.06it/s] 8%|▊ | 900/10833 [00:11<01:17, 128.15it/s] 8%|▊ | 913/10833 [00:11<01:17, 128.09it/s] 9%|▊ | 926/10833 [00:11<01:17, 128.11it/s] 9%|▊ | 939/10833 [00:11<01:17, 127.80it/s] 9%|▉ | 952/10833 [00:11<01:17, 127.95it/s] 9%|▉ | 965/10833 [00:11<01:17, 127.99it/s] 9%|▉ | 978/10833 [00:11<01:17, 127.04it/s] 9%|▉ | 991/10833 [00:11<01:17, 127.45it/s] 9%|▉ | 1004/10833 [00:12<01:16, 127.73it/s] 9%|▉ | 1017/10833 [00:12<01:16, 127.89it/s] 10%|▉ | 1030/10833 [00:12<01:16, 128.04it/s] 10%|▉ | 1043/10833 [00:12<01:16, 128.15it/s] 10%|▉ | 1056/10833 [00:12<01:16, 128.18it/s] 10%|▉ | 1069/10833 [00:12<01:16, 128.29it/s] 10%|▉ | 1082/10833 [00:12<01:15, 128.40it/s] 10%|█ | 1095/10833 [00:12<01:15, 128.48it/s] 10%|█ | 1108/10833 [00:12<01:15, 128.06it/s] 10%|█ | 1121/10833 [00:12<01:15, 128.19it/s] 10%|█ | 1134/10833 [00:13<01:15, 128.25it/s] 11%|█ | 1147/10833 [00:13<01:16, 127.25it/s] 11%|█ | 1160/10833 [00:13<01:15, 127.57it/s] 11%|█ | 1173/10833 [00:13<01:15, 127.75it/s] 11%|█ | 1186/10833 [00:13<01:15, 127.92it/s] 11%|█ | 1199/10833 [00:13<01:15, 128.14it/s] 11%|█ | 1212/10833 [00:13<01:15, 128.23it/s] 11%|█▏ | 1225/10833 [00:13<01:14, 128.37it/s] 11%|█▏ | 1238/10833 [00:13<01:14, 128.31it/s] 12%|█▏ | 1251/10833 [00:14<01:14, 128.33it/s] 12%|█▏ | 1264/10833 [00:14<01:14, 128.33it/s] 12%|█▏ | 1277/10833 [00:14<01:14, 127.99it/s] 12%|█▏ | 1290/10833 [00:14<01:14, 128.10it/s] 12%|█▏ | 1303/10833 [00:14<01:14, 128.24it/s] 12%|█▏ | 1316/10833 [00:14<01:14, 128.12it/s] 12%|█▏ | 1329/10833 [00:14<01:14, 127.32it/s] 12%|█▏ | 1342/10833 [00:14<01:14, 127.69it/s] 13%|█▎ | 1355/10833 [00:14<01:14, 127.85it/s] 13%|█▎ | 1368/10833 [00:14<01:14, 127.89it/s] 13%|█▎ | 1381/10833 [00:15<01:13, 128.00it/s] 13%|█▎ | 1394/10833 [00:15<01:13, 128.17it/s] 13%|█▎ | 1407/10833 [00:15<01:13, 128.20it/s] 13%|█▎ | 1420/10833 [00:15<01:13, 128.32it/s] 13%|█▎ | 1433/10833 [00:15<01:13, 128.32it/s] 13%|█▎ | 1446/10833 [00:15<01:13, 128.28it/s] 13%|█▎ | 1459/10833 [00:15<01:14, 125.95it/s] 14%|█▎ | 1472/10833 [00:15<01:13, 126.78it/s] 14%|█▎ | 1485/10833 [00:15<01:13, 127.34it/s] 14%|█▍ | 1498/10833 [00:15<01:13, 126.64it/s] 14%|█▍ | 1511/10833 [00:16<01:13, 127.24it/s] 14%|█▍ | 1524/10833 [00:16<01:12, 127.57it/s] 14%|█▍ | 1537/10833 [00:16<01:12, 127.82it/s] 14%|█▍ | 1550/10833 [00:16<01:12, 127.97it/s] 14%|█▍ | 1563/10833 [00:16<01:12, 128.10it/s] 15%|█▍ | 1576/10833 [00:16<01:12, 128.24it/s] 15%|█▍ | 1589/10833 [00:16<01:12, 128.23it/s] 15%|█▍ | 1602/10833 [00:16<01:11, 128.24it/s] 15%|█▍ | 1615/10833 [00:16<01:11, 128.35it/s] 15%|█▌ | 1628/10833 [00:16<01:11, 127.95it/s] 15%|█▌ | 1641/10833 [00:17<01:11, 128.10it/s] 15%|█▌ | 1654/10833 [00:17<01:11, 128.17it/s] 15%|█▌ | 1667/10833 [00:17<01:12, 127.30it/s] 16%|█▌ | 1680/10833 [00:17<01:11, 127.61it/s] 16%|█▌ | 1693/10833 [00:17<01:11, 127.89it/s] 16%|█▌ | 1706/10833 [00:17<01:11, 127.95it/s] 16%|█▌ | 1719/10833 [00:17<01:11, 128.10it/s] 16%|█▌ | 1732/10833 [00:17<01:11, 128.18it/s] 16%|█▌ | 1745/10833 [00:17<01:10, 128.28it/s] 16%|█▌ | 1758/10833 [00:17<01:10, 128.33it/s] 16%|█▋ | 1771/10833 [00:18<01:10, 128.37it/s] 16%|█▋ | 1784/10833 [00:18<01:10, 128.31it/s] 17%|█▋ | 1797/10833 [00:18<01:10, 128.05it/s] 17%|█▋ | 1810/10833 [00:18<01:10, 128.08it/s] 17%|█▋ | 1823/10833 [00:18<01:10, 128.24it/s] 17%|█▋ | 1836/10833 [00:18<01:10, 127.39it/s] 17%|█▋ | 1849/10833 [00:18<01:10, 127.76it/s] 17%|█▋ | 1862/10833 [00:18<01:10, 128.02it/s] 17%|█▋ | 1875/10833 [00:18<01:09, 128.11it/s] 17%|█▋ | 1888/10833 [00:18<01:09, 128.17it/s] 18%|█▊ | 1901/10833 [00:19<01:09, 128.20it/s] 18%|█▊ | 1914/10833 [00:19<01:09, 128.25it/s] 18%|█▊ | 1927/10833 [00:19<01:09, 128.18it/s] 18%|█▊ | 1940/10833 [00:19<01:09, 128.19it/s] 18%|█▊ | 1953/10833 [00:19<01:09, 128.33it/s] 18%|█▊ | 1966/10833 [00:19<01:09, 128.36it/s] 18%|█▊ | 1979/10833 [00:19<01:09, 128.03it/s] 18%|█▊ | 1992/10833 [00:19<01:08, 128.24it/s] 19%|█▊ | 2005/10833 [00:19<01:09, 127.17it/s] 19%|█▊ | 2018/10833 [00:20<01:09, 127.46it/s] 19%|█▊ | 2031/10833 [00:20<01:08, 127.72it/s] 19%|█▉ | 2044/10833 [00:20<01:08, 127.91it/s] 19%|█▉ | 2057/10833 [00:20<01:08, 128.00it/s] 19%|█▉ | 2070/10833 [00:20<01:08, 128.09it/s] 19%|█▉ | 2083/10833 [00:20<01:08, 128.18it/s] 19%|█▉ | 2096/10833 [00:20<01:08, 128.26it/s] 19%|█▉ | 2109/10833 [00:20<01:08, 128.14it/s] 20%|█▉ | 2122/10833 [00:20<01:07, 128.20it/s] 20%|█▉ | 2135/10833 [00:20<01:07, 128.32it/s] 20%|█▉ | 2148/10833 [00:21<01:07, 128.08it/s] 20%|█▉ | 2161/10833 [00:21<01:07, 128.15it/s] 20%|██ | 2174/10833 [00:21<01:07, 128.19it/s] 20%|██ | 2187/10833 [00:21<01:07, 127.36it/s] 20%|██ | 2200/10833 [00:21<01:07, 127.63it/s] 20%|██ | 2213/10833 [00:21<01:07, 127.85it/s] 21%|██ | 2226/10833 [00:21<01:07, 128.00it/s] 21%|██ | 2239/10833 [00:21<01:07, 128.16it/s] 21%|██ | 2252/10833 [00:21<01:06, 128.20it/s] 21%|██ | 2265/10833 [00:21<01:06, 128.20it/s] 21%|██ | 2278/10833 [00:22<01:06, 128.15it/s] 21%|██ | 2291/10833 [00:22<01:06, 128.19it/s] 21%|██▏ | 2304/10833 [00:22<01:06, 128.21it/s] 21%|██▏ | 2317/10833 [00:22<01:06, 127.79it/s] 22%|██▏ | 2330/10833 [00:22<01:06, 127.82it/s] 22%|██▏ | 2343/10833 [00:22<01:06, 128.00it/s] 22%|██▏ | 2356/10833 [00:22<01:06, 127.31it/s] 22%|██▏ | 2369/10833 [00:22<01:06, 127.62it/s] 22%|██▏ | 2382/10833 [00:22<01:06, 127.87it/s] 22%|██▏ | 2395/10833 [00:22<01:05, 128.03it/s] 22%|██▏ | 2408/10833 [00:23<01:05, 128.11it/s] 22%|██▏ | 2421/10833 [00:23<01:05, 128.11it/s] 22%|██▏ | 2434/10833 [00:23<01:05, 128.12it/s] 23%|██▎ | 2447/10833 [00:23<01:05, 128.18it/s] 23%|██▎ | 2460/10833 [00:23<01:05, 128.18it/s] 23%|██▎ | 2473/10833 [00:23<01:05, 128.17it/s] 23%|██▎ | 2486/10833 [00:23<01:05, 128.25it/s] 23%|██▎ | 2499/10833 [00:23<01:05, 128.02it/s] 23%|██▎ | 2512/10833 [00:23<01:04, 128.17it/s] 23%|██▎ | 2525/10833 [00:23<01:05, 127.20it/s] 23%|██▎ | 2538/10833 [00:24<01:05, 127.51it/s] 24%|██▎ | 2551/10833 [00:24<01:04, 127.69it/s] 24%|██▎ | 2564/10833 [00:24<01:04, 127.86it/s] 24%|██▍ | 2577/10833 [00:24<01:04, 128.00it/s] 24%|██▍ | 2590/10833 [00:24<01:04, 127.97it/s] 24%|██▍ | 2603/10833 [00:24<01:04, 128.19it/s] 24%|██▍ | 2616/10833 [00:24<01:04, 128.26it/s] 24%|██▍ | 2629/10833 [00:24<01:03, 128.25it/s] 24%|██▍ | 2642/10833 [00:24<01:03, 128.26it/s] 25%|██▍ | 2655/10833 [00:24<01:03, 128.27it/s] 25%|██▍ | 2668/10833 [00:25<01:03, 127.92it/s] 25%|██▍ | 2681/10833 [00:25<01:03, 128.05it/s] 25%|██▍ | 2694/10833 [00:25<01:04, 127.16it/s] 25%|██▍ | 2707/10833 [00:25<01:03, 127.56it/s] 25%|██▌ | 2720/10833 [00:25<01:03, 127.69it/s] 25%|██▌ | 2733/10833 [00:25<01:03, 127.86it/s] 25%|██▌ | 2746/10833 [00:25<01:03, 127.90it/s] 25%|██▌ | 2759/10833 [00:25<01:03, 128.03it/s] 26%|██▌ | 2772/10833 [00:25<01:02, 128.08it/s] 26%|██▌ | 2785/10833 [00:25<01:02, 128.10it/s] 26%|██▌ | 2798/10833 [00:26<01:02, 128.19it/s] 26%|██▌ | 2811/10833 [00:26<01:02, 128.17it/s] 26%|██▌ | 2824/10833 [00:26<01:02, 127.95it/s] 26%|██▌ | 2837/10833 [00:26<01:02, 127.66it/s] 26%|██▋ | 2850/10833 [00:26<01:02, 127.95it/s] 26%|██▋ | 2863/10833 [00:26<01:02, 127.18it/s] 27%|██▋ | 2876/10833 [00:26<01:02, 127.33it/s] 27%|██▋ | 2889/10833 [00:26<01:02, 127.54it/s] 27%|██▋ | 2902/10833 [00:26<01:02, 127.77it/s] 27%|██▋ | 2915/10833 [00:27<01:01, 127.81it/s] 27%|██▋ | 2928/10833 [00:27<01:01, 127.93it/s] 27%|██▋ | 2941/10833 [00:27<01:01, 128.04it/s] 27%|██▋ | 2954/10833 [00:27<01:01, 128.14it/s] 27%|██▋ | 2967/10833 [00:27<01:01, 128.23it/s] 28%|██▊ | 2980/10833 [00:27<01:01, 128.27it/s] 28%|██▊ | 2993/10833 [00:27<01:01, 128.30it/s] 28%|██▊ | 3006/10833 [00:27<01:00, 128.40it/s] 28%|██▊ | 3019/10833 [00:27<01:01, 127.88it/s] 28%|██▊ | 3032/10833 [00:27<01:00, 128.01it/s] 28%|██▊ | 3045/10833 [00:28<01:01, 127.10it/s] 28%|██▊ | 3058/10833 [00:28<01:00, 127.49it/s] 28%|██▊ | 3071/10833 [00:28<01:00, 127.70it/s] 28%|██▊ | 3084/10833 [00:28<01:00, 127.91it/s] 29%|██▊ | 3097/10833 [00:28<01:00, 128.11it/s] 29%|██▊ | 3110/10833 [00:28<01:00, 128.20it/s] 29%|██▉ | 3123/10833 [00:28<01:00, 128.29it/s] 29%|██▉ | 3136/10833 [00:28<00:59, 128.34it/s] 29%|██▉ | 3149/10833 [00:28<00:59, 128.26it/s] 29%|██▉ | 3162/10833 [00:28<00:59, 128.27it/s] 29%|██▉ | 3175/10833 [00:29<00:59, 128.23it/s] 29%|██▉ | 3188/10833 [00:29<00:59, 127.74it/s] 30%|██▉ | 3201/10833 [00:29<00:59, 127.95it/s] 30%|██▉ | 3214/10833 [00:29<00:59, 127.02it/s] 30%|██▉ | 3227/10833 [00:29<00:59, 127.39it/s] 30%|██▉ | 3240/10833 [00:29<00:59, 127.66it/s] 30%|███ | 3253/10833 [00:29<00:59, 127.95it/s] 30%|███ | 3266/10833 [00:29<00:59, 128.01it/s] 30%|███ | 3279/10833 [00:29<00:58, 128.08it/s] 30%|███ | 3292/10833 [00:29<00:58, 128.09it/s] 31%|███ | 3305/10833 [00:30<00:58, 128.18it/s] 31%|███ | 3318/10833 [00:30<00:58, 128.23it/s] 31%|███ | 3331/10833 [00:30<00:58, 128.21it/s] 31%|███ | 3344/10833 [00:30<00:58, 128.20it/s] 31%|███ | 3357/10833 [00:30<00:58, 127.83it/s] 31%|███ | 3370/10833 [00:30<00:58, 128.04it/s] 31%|███ | 3383/10833 [00:30<00:58, 127.23it/s] 31%|███▏ | 3396/10833 [00:30<00:58, 127.69it/s] 31%|███▏ | 3409/10833 [00:30<00:58, 127.98it/s] 32%|███▏ | 3422/10833 [00:30<00:57, 128.08it/s] 32%|███▏ | 3435/10833 [00:31<00:57, 128.08it/s] 32%|███▏ | 3448/10833 [00:31<00:57, 128.06it/s] 32%|███▏ | 3461/10833 [00:31<00:57, 128.03it/s] 32%|███▏ | 3474/10833 [00:31<00:57, 128.08it/s] 32%|███▏ | 3487/10833 [00:31<00:57, 128.12it/s] 32%|███▏ | 3500/10833 [00:31<00:57, 128.17it/s] 32%|███▏ | 3513/10833 [00:31<00:57, 128.28it/s] 33%|███▎ | 3526/10833 [00:31<00:58, 124.68it/s] 33%|███▎ | 3539/10833 [00:31<00:58, 125.75it/s] 33%|███▎ | 3552/10833 [00:31<00:57, 125.71it/s] 33%|███▎ | 3565/10833 [00:32<00:57, 126.43it/s] 33%|███▎ | 3578/10833 [00:32<00:57, 126.90it/s] 33%|███▎ | 3591/10833 [00:32<00:56, 127.17it/s] 33%|███▎ | 3604/10833 [00:32<00:56, 127.46it/s] 33%|███▎ | 3617/10833 [00:32<00:56, 127.77it/s] 34%|███▎ | 3630/10833 [00:32<00:56, 127.98it/s] 34%|███▎ | 3643/10833 [00:32<00:56, 128.08it/s] 34%|███▎ | 3656/10833 [00:32<00:55, 128.20it/s] 34%|███▍ | 3669/10833 [00:32<00:55, 128.22it/s] 34%|███▍ | 3682/10833 [00:33<00:55, 128.24it/s] 34%|███▍ | 3695/10833 [00:33<00:55, 128.26it/s] 34%|███▍ | 3708/10833 [00:33<00:55, 127.88it/s] 34%|███▍ | 3721/10833 [00:33<00:55, 127.20it/s] 34%|███▍ | 3734/10833 [00:33<00:55, 127.49it/s] 35%|███▍ | 3747/10833 [00:33<00:55, 127.80it/s] 35%|███▍ | 3760/10833 [00:33<00:55, 127.91it/s] 35%|███▍ | 3773/10833 [00:33<00:55, 128.01it/s] 35%|███▍ | 3786/10833 [00:33<00:55, 128.12it/s] 35%|███▌ | 3799/10833 [00:33<00:54, 128.20it/s] 35%|███▌ | 3812/10833 [00:34<00:54, 128.12it/s] 35%|███▌ | 3825/10833 [00:34<00:54, 127.92it/s] 35%|███▌ | 3838/10833 [00:34<00:54, 128.04it/s] 36%|███▌ | 3851/10833 [00:34<00:54, 128.03it/s] 36%|███▌ | 3864/10833 [00:34<00:54, 128.07it/s] 36%|███▌ | 3877/10833 [00:34<00:54, 127.84it/s] 36%|███▌ | 3890/10833 [00:34<00:54, 127.90it/s] 36%|███▌ | 3903/10833 [00:34<00:54, 127.06it/s] 36%|███▌ | 3916/10833 [00:34<00:54, 127.35it/s] 36%|███▋ | 3929/10833 [00:34<00:54, 127.66it/s] 36%|███▋ | 3942/10833 [00:35<00:53, 127.85it/s] 37%|███▋ | 3955/10833 [00:35<00:53, 128.04it/s] 37%|███▋ | 3968/10833 [00:35<00:53, 128.04it/s] 37%|███▋ | 3981/10833 [00:35<00:53, 128.00it/s] 37%|███▋ | 3994/10833 [00:35<00:53, 128.02it/s] 37%|███▋ | 4007/10833 [00:35<00:53, 128.06it/s] 37%|███▋ | 4020/10833 [00:35<00:53, 128.18it/s] 37%|███▋ | 4033/10833 [00:35<00:53, 128.12it/s] 37%|███▋ | 4046/10833 [00:35<00:53, 127.74it/s] 37%|███▋ | 4059/10833 [00:35<00:53, 127.79it/s] 38%|███▊ | 4072/10833 [00:36<00:53, 127.02it/s] 38%|███▊ | 4085/10833 [00:36<00:53, 127.30it/s] 38%|███▊ | 4098/10833 [00:36<00:52, 127.59it/s] 38%|███▊ | 4111/10833 [00:36<00:52, 127.77it/s] 38%|███▊ | 4124/10833 [00:36<00:52, 127.85it/s] 38%|███▊ | 4137/10833 [00:36<00:52, 127.67it/s] 38%|███▊ | 4150/10833 [00:36<00:52, 127.92it/s] 38%|███▊ | 4163/10833 [00:36<00:52, 128.02it/s] 39%|███▊ | 4176/10833 [00:36<00:51, 128.13it/s] 39%|███▊ | 4189/10833 [00:36<00:51, 128.25it/s] 39%|███▉ | 4202/10833 [00:37<00:51, 128.33it/s] 39%|███▉ | 4215/10833 [00:37<00:51, 128.29it/s] 39%|███▉ | 4228/10833 [00:37<00:51, 127.89it/s] 39%|███▉ | 4241/10833 [00:37<00:51, 126.83it/s] 39%|███▉ | 4254/10833 [00:37<00:51, 127.32it/s] 39%|███▉ | 4267/10833 [00:37<00:51, 127.63it/s] 40%|███▉ | 4280/10833 [00:37<00:51, 127.88it/s] 40%|███▉ | 4293/10833 [00:37<00:51, 127.97it/s] 40%|███▉ | 4306/10833 [00:37<00:50, 128.07it/s] 40%|███▉ | 4319/10833 [00:37<00:50, 128.09it/s] 40%|███▉ | 4332/10833 [00:38<00:50, 128.12it/s] 40%|████ | 4345/10833 [00:38<00:50, 128.12it/s] 40%|████ | 4358/10833 [00:38<00:50, 128.15it/s] 40%|████ | 4371/10833 [00:38<00:50, 128.27it/s] 40%|████ | 4384/10833 [00:38<00:50, 128.27it/s] 41%|████ | 4397/10833 [00:38<00:50, 128.04it/s] 41%|████ | 4410/10833 [00:38<00:50, 127.33it/s] 41%|████ | 4423/10833 [00:38<00:50, 127.71it/s] 41%|████ | 4436/10833 [00:38<00:50, 127.87it/s] 41%|████ | 4449/10833 [00:39<00:49, 127.89it/s] 41%|████ | 4462/10833 [00:39<00:49, 128.03it/s] 41%|████▏ | 4475/10833 [00:39<00:49, 128.03it/s] 41%|████▏ | 4488/10833 [00:39<00:49, 128.10it/s] 42%|████▏ | 4501/10833 [00:39<00:49, 128.08it/s] 42%|████▏ | 4514/10833 [00:39<00:49, 128.14it/s] 42%|████▏ | 4527/10833 [00:39<00:49, 128.19it/s] 42%|████▏ | 4540/10833 [00:39<00:49, 128.26it/s] 42%|████▏ | 4553/10833 [00:39<00:48, 128.19it/s] 42%|████▏ | 4566/10833 [00:39<00:49, 127.87it/s] 42%|████▏ | 4579/10833 [00:40<00:49, 126.86it/s] 42%|████▏ | 4592/10833 [00:40<00:49, 127.21it/s] 43%|████▎ | 4605/10833 [00:40<00:48, 127.51it/s] 43%|████▎ | 4618/10833 [00:40<00:48, 127.61it/s] 43%|████▎ | 4631/10833 [00:40<00:48, 127.81it/s] 43%|████▎ | 4644/10833 [00:40<00:48, 127.92it/s] 43%|████▎ | 4657/10833 [00:40<00:48, 128.01it/s] 43%|████▎ | 4670/10833 [00:40<00:48, 128.02it/s] 43%|████▎ | 4683/10833 [00:40<00:48, 128.04it/s] 43%|████▎ | 4696/10833 [00:40<00:47, 128.11it/s] 43%|████▎ | 4709/10833 [00:41<00:47, 128.21it/s] 44%|████▎ | 4722/10833 [00:41<00:47, 128.17it/s] 44%|████▎ | 4735/10833 [00:41<00:47, 128.13it/s] 44%|████▍ | 4748/10833 [00:41<00:47, 127.07it/s] 44%|████▍ | 4761/10833 [00:41<00:47, 127.28it/s] 44%|████▍ | 4774/10833 [00:41<00:47, 127.54it/s] 44%|████▍ | 4787/10833 [00:41<00:47, 127.66it/s] 44%|████▍ | 4800/10833 [00:41<00:47, 127.82it/s] 44%|████▍ | 4813/10833 [00:41<00:47, 127.95it/s] 45%|████▍ | 4826/10833 [00:41<00:46, 128.07it/s] 45%|████▍ | 4839/10833 [00:42<00:46, 128.03it/s] 45%|████▍ | 4852/10833 [00:42<00:46, 128.12it/s] 45%|████▍ | 4865/10833 [00:42<00:46, 128.07it/s] 45%|████▌ | 4878/10833 [00:42<00:46, 128.13it/s] 45%|████▌ | 4891/10833 [00:42<00:46, 128.06it/s] 45%|████▌ | 4904/10833 [00:42<00:46, 128.11it/s] 45%|████▌ | 4917/10833 [00:42<00:46, 127.71it/s] 46%|████▌ | 4930/10833 [00:42<00:46, 126.99it/s] 46%|████▌ | 4943/10833 [00:42<00:46, 127.34it/s] 46%|████▌ | 4956/10833 [00:42<00:46, 127.57it/s] 46%|████▌ | 4969/10833 [00:43<00:45, 127.67it/s] 46%|████▌ | 4982/10833 [00:43<00:45, 127.88it/s] 46%|████▌ | 4995/10833 [00:43<00:45, 127.98it/s] 46%|████▌ | 5008/10833 [00:43<00:45, 128.09it/s] 46%|████▋ | 5021/10833 [00:43<00:45, 128.16it/s] 46%|████▋ | 5034/10833 [00:43<00:45, 128.26it/s] 47%|████▋ | 5047/10833 [00:43<00:45, 128.27it/s] 47%|████▋ | 5060/10833 [00:43<00:45, 128.28it/s] 47%|████▋ | 5073/10833 [00:43<00:44, 128.31it/s] 47%|████▋ | 5086/10833 [00:43<00:44, 128.00it/s] 47%|████▋ | 5099/10833 [00:44<00:45, 126.96it/s] 47%|████▋ | 5112/10833 [00:44<00:44, 127.34it/s] 47%|████▋ | 5125/10833 [00:44<00:44, 127.49it/s] 47%|████▋ | 5138/10833 [00:44<00:44, 127.71it/s] 48%|████▊ | 5151/10833 [00:44<00:44, 127.89it/s] 48%|████▊ | 5164/10833 [00:44<00:44, 128.02it/s] 48%|████▊ | 5177/10833 [00:44<00:44, 128.10it/s] 48%|████▊ | 5190/10833 [00:44<00:44, 128.09it/s] 48%|████▊ | 5203/10833 [00:44<00:43, 128.15it/s] 48%|████▊ | 5216/10833 [00:45<00:43, 128.19it/s] 48%|████▊ | 5229/10833 [00:45<00:43, 128.06it/s] 48%|████▊ | 5242/10833 [00:45<00:43, 127.97it/s] 49%|████▊ | 5255/10833 [00:45<00:43, 127.91it/s] 49%|████▊ | 5268/10833 [00:45<00:43, 126.85it/s] 49%|████▊ | 5281/10833 [00:45<00:43, 127.13it/s] 49%|████▉ | 5294/10833 [00:45<00:43, 127.39it/s] 49%|████▉ | 5307/10833 [00:45<00:43, 127.62it/s] 49%|████▉ | 5320/10833 [00:45<00:43, 127.76it/s] 49%|████▉ | 5333/10833 [00:45<00:42, 127.93it/s] 49%|████▉ | 5346/10833 [00:46<00:42, 128.03it/s] 49%|████▉ | 5359/10833 [00:46<00:42, 128.13it/s] 50%|████▉ | 5372/10833 [00:46<00:42, 128.02it/s] 50%|████▉ | 5385/10833 [00:46<00:42, 127.92it/s] 50%|████▉ | 5398/10833 [00:46<00:42, 127.89it/s] 50%|████▉ | 5411/10833 [00:46<00:42, 128.05it/s] 50%|█████ | 5424/10833 [00:46<00:42, 128.09it/s] 50%|█████ | 5437/10833 [00:46<00:42, 126.89it/s] 50%|█████ | 5450/10833 [00:46<00:42, 127.33it/s] 50%|█████ | 5463/10833 [00:46<00:42, 127.59it/s] 51%|█████ | 5476/10833 [00:47<00:41, 127.58it/s] 51%|█████ | 5489/10833 [00:47<00:41, 127.77it/s] 51%|█████ | 5502/10833 [00:47<00:41, 127.90it/s] 51%|█████ | 5515/10833 [00:47<00:41, 128.00it/s] 51%|█████ | 5528/10833 [00:47<00:41, 127.90it/s] 51%|█████ | 5541/10833 [00:47<00:41, 128.06it/s] 51%|█████▏ | 5554/10833 [00:47<00:41, 128.21it/s] 51%|█████▏ | 5567/10833 [00:47<00:41, 128.16it/s] 52%|█████▏ | 5580/10833 [00:47<00:40, 128.21it/s] 52%|█████▏ | 5593/10833 [00:47<00:40, 128.21it/s] 52%|█████▏ | 5606/10833 [00:48<00:42, 123.82it/s] 52%|█████▏ | 5619/10833 [00:48<00:41, 125.06it/s] 52%|█████▏ | 5632/10833 [00:48<00:41, 125.97it/s] 52%|█████▏ | 5645/10833 [00:48<00:40, 126.62it/s] 52%|█████▏ | 5658/10833 [00:48<00:40, 127.04it/s] 52%|█████▏ | 5671/10833 [00:48<00:40, 127.41it/s] 52%|█████▏ | 5684/10833 [00:48<00:40, 127.71it/s] 53%|█████▎ | 5697/10833 [00:48<00:40, 127.90it/s] 53%|█████▎ | 5710/10833 [00:48<00:40, 128.06it/s] 53%|█████▎ | 5723/10833 [00:48<00:39, 128.09it/s] 53%|█████▎ | 5736/10833 [00:49<00:39, 128.12it/s] 53%|█████▎ | 5749/10833 [00:49<00:39, 128.17it/s] 53%|█████▎ | 5762/10833 [00:49<00:39, 128.03it/s] 53%|█████▎ | 5775/10833 [00:49<00:39, 127.68it/s] 53%|█████▎ | 5788/10833 [00:49<00:39, 126.91it/s] 54%|█████▎ | 5801/10833 [00:49<00:39, 127.37it/s] 54%|█████▎ | 5814/10833 [00:49<00:39, 127.64it/s] 54%|█████▍ | 5827/10833 [00:49<00:39, 127.79it/s] 54%|█████▍ | 5840/10833 [00:49<00:39, 127.89it/s] 54%|█████▍ | 5853/10833 [00:50<00:38, 128.03it/s] 54%|█████▍ | 5866/10833 [00:50<00:38, 128.03it/s] 54%|█████▍ | 5879/10833 [00:50<00:38, 128.03it/s] 54%|█████▍ | 5892/10833 [00:50<00:38, 128.06it/s] 55%|█████▍ | 5905/10833 [00:50<00:38, 128.05it/s] 55%|█████▍ | 5918/10833 [00:50<00:38, 128.02it/s] 55%|█████▍ | 5931/10833 [00:50<00:38, 128.07it/s] 55%|█████▍ | 5944/10833 [00:50<00:38, 128.14it/s] 55%|█████▍ | 5957/10833 [00:50<00:38, 127.12it/s] 55%|█████▌ | 5970/10833 [00:50<00:38, 127.46it/s] 55%|█████▌ | 5983/10833 [00:51<00:37, 127.67it/s] 55%|█████▌ | 5996/10833 [00:51<00:37, 127.72it/s] 55%|█████▌ | 6009/10833 [00:51<00:37, 127.73it/s] 56%|█████▌ | 6022/10833 [00:51<00:37, 127.84it/s] 56%|█████▌ | 6035/10833 [00:51<00:37, 127.87it/s] 56%|█████▌ | 6048/10833 [00:51<00:37, 127.94it/s] 56%|█████▌ | 6061/10833 [00:51<00:37, 127.93it/s] 56%|█████▌ | 6074/10833 [00:51<00:37, 128.03it/s] 56%|█████▌ | 6087/10833 [00:51<00:37, 127.98it/s] 56%|█████▋ | 6100/10833 [00:51<00:36, 128.12it/s] 56%|█████▋ | 6113/10833 [00:52<00:36, 128.01it/s] 57%|█████▋ | 6126/10833 [00:52<00:37, 126.89it/s] 57%|█████▋ | 6139/10833 [00:52<00:36, 127.20it/s] 57%|█████▋ | 6152/10833 [00:52<00:36, 127.50it/s] 57%|█████▋ | 6165/10833 [00:52<00:36, 127.68it/s] 57%|█████▋ | 6178/10833 [00:52<00:36, 127.77it/s] 57%|█████▋ | 6191/10833 [00:52<00:36, 127.97it/s] 57%|█████▋ | 6204/10833 [00:52<00:36, 128.03it/s] 57%|█████▋ | 6217/10833 [00:52<00:36, 128.03it/s] 58%|█████▊ | 6230/10833 [00:52<00:35, 127.98it/s] 58%|█████▊ | 6243/10833 [00:53<00:35, 127.97it/s] 58%|█████▊ | 6256/10833 [00:53<00:35, 128.03it/s] 58%|█████▊ | 6269/10833 [00:53<00:35, 128.04it/s] 58%|█████▊ | 6282/10833 [00:53<00:35, 127.98it/s] 58%|█████▊ | 6295/10833 [00:53<00:35, 126.65it/s] 58%|█████▊ | 6308/10833 [00:53<00:35, 127.09it/s] 58%|█████▊ | 6321/10833 [00:53<00:35, 127.38it/s] 58%|█████▊ | 6334/10833 [00:53<00:35, 127.62it/s] 59%|█████▊ | 6347/10833 [00:53<00:35, 127.86it/s] 59%|█████▊ | 6360/10833 [00:53<00:34, 127.94it/s] 59%|█████▉ | 6373/10833 [00:54<00:34, 127.98it/s] 59%|█████▉ | 6386/10833 [00:54<00:34, 128.09it/s] 59%|█████▉ | 6399/10833 [00:54<00:34, 128.05it/s] 59%|█████▉ | 6412/10833 [00:54<00:34, 128.13it/s] 59%|█████▉ | 6425/10833 [00:54<00:34, 128.20it/s] 59%|█████▉ | 6438/10833 [00:54<00:34, 128.16it/s] 60%|█████▉ | 6451/10833 [00:54<00:34, 128.30it/s] 60%|█████▉ | 6464/10833 [00:54<00:34, 127.26it/s] 60%|█████▉ | 6477/10833 [00:54<00:34, 127.21it/s] 60%|█████▉ | 6490/10833 [00:54<00:34, 127.55it/s] 60%|██████ | 6503/10833 [00:55<00:33, 127.62it/s] 60%|██████ | 6516/10833 [00:55<00:33, 127.71it/s] 60%|██████ | 6529/10833 [00:55<00:33, 127.82it/s] 60%|██████ | 6542/10833 [00:55<00:33, 127.88it/s] 61%|██████ | 6555/10833 [00:55<00:33, 127.88it/s] 61%|██████ | 6568/10833 [00:55<00:33, 127.96it/s] 61%|██████ | 6581/10833 [00:55<00:33, 128.04it/s] 61%|██████ | 6594/10833 [00:55<00:33, 128.08it/s] 61%|██████ | 6607/10833 [00:55<00:32, 128.07it/s] 61%|██████ | 6620/10833 [00:56<00:32, 128.19it/s] 61%|██████ | 6633/10833 [00:56<00:32, 127.31it/s] 61%|██████▏ | 6646/10833 [00:56<00:32, 127.19it/s] 61%|██████▏ | 6659/10833 [00:56<00:32, 127.39it/s] 62%|██████▏ | 6672/10833 [00:56<00:32, 127.51it/s] 62%|██████▏ | 6685/10833 [00:56<00:32, 127.63it/s] 62%|██████▏ | 6698/10833 [00:56<00:32, 127.81it/s] 62%|██████▏ | 6711/10833 [00:56<00:32, 127.95it/s] 62%|██████▏ | 6724/10833 [00:56<00:32, 128.04it/s] 62%|██████▏ | 6737/10833 [00:56<00:31, 128.08it/s] 62%|██████▏ | 6750/10833 [00:57<00:31, 128.17it/s] 62%|██████▏ | 6763/10833 [00:57<00:31, 128.18it/s] 63%|██████▎ | 6776/10833 [00:57<00:31, 128.20it/s] 63%|██████▎ | 6789/10833 [00:57<00:31, 128.13it/s] 63%|██████▎ | 6802/10833 [00:57<00:31, 128.02it/s] 63%|██████▎ | 6815/10833 [00:57<00:31, 126.90it/s] 63%|██████▎ | 6828/10833 [00:57<00:31, 127.34it/s] 63%|██████▎ | 6841/10833 [00:57<00:31, 127.62it/s] 63%|██████▎ | 6854/10833 [00:57<00:31, 127.80it/s] 63%|██████▎ | 6867/10833 [00:57<00:31, 127.87it/s] 64%|██████▎ | 6880/10833 [00:58<00:30, 127.92it/s] 64%|██████▎ | 6893/10833 [00:58<00:30, 127.98it/s] 64%|██████▎ | 6906/10833 [00:58<00:30, 128.05it/s] 64%|██████▍ | 6919/10833 [00:58<00:30, 128.14it/s] 64%|██████▍ | 6932/10833 [00:58<00:30, 128.07it/s] 64%|██████▍ | 6945/10833 [00:58<00:30, 127.92it/s] 64%|██████▍ | 6958/10833 [00:58<00:30, 127.84it/s] 64%|██████▍ | 6971/10833 [00:58<00:30, 127.94it/s] 64%|██████▍ | 6984/10833 [00:58<00:30, 127.19it/s] 65%|██████▍ | 6997/10833 [00:58<00:30, 127.14it/s] 65%|██████▍ | 7010/10833 [00:59<00:30, 127.34it/s] 65%|██████▍ | 7023/10833 [00:59<00:29, 127.60it/s] 65%|██████▍ | 7036/10833 [00:59<00:29, 127.74it/s] 65%|██████▌ | 7049/10833 [00:59<00:29, 127.83it/s] 65%|██████▌ | 7062/10833 [00:59<00:29, 127.97it/s] 65%|██████▌ | 7075/10833 [00:59<00:29, 128.05it/s] 65%|██████▌ | 7088/10833 [00:59<00:29, 128.05it/s] 66%|██████▌ | 7101/10833 [00:59<00:29, 127.98it/s] 66%|██████▌ | 7114/10833 [00:59<00:29, 128.15it/s] 66%|██████▌ | 7127/10833 [00:59<00:28, 128.16it/s] 66%|██████▌ | 7140/10833 [01:00<00:28, 128.07it/s] 66%|██████▌ | 7153/10833 [01:00<00:28, 127.05it/s] 66%|██████▌ | 7166/10833 [01:00<00:28, 126.99it/s] 66%|██████▋ | 7179/10833 [01:00<00:28, 127.30it/s] 66%|██████▋ | 7192/10833 [01:00<00:28, 127.49it/s] 67%|██████▋ | 7205/10833 [01:00<00:28, 127.72it/s] 67%|██████▋ | 7218/10833 [01:00<00:28, 127.94it/s] 67%|██████▋ | 7231/10833 [01:00<00:28, 127.95it/s] 67%|██████▋ | 7244/10833 [01:00<00:28, 128.17it/s] 67%|██████▋ | 7257/10833 [01:00<00:27, 128.20it/s] 67%|██████▋ | 7270/10833 [01:01<00:27, 128.09it/s] 67%|██████▋ | 7283/10833 [01:01<00:27, 127.98it/s] 67%|██████▋ | 7296/10833 [01:01<00:27, 127.96it/s] 67%|██████▋ | 7309/10833 [01:01<00:27, 127.94it/s] 68%|██████▊ | 7322/10833 [01:01<00:27, 126.99it/s] 68%|██████▊ | 7335/10833 [01:01<00:27, 127.01it/s] 68%|██████▊ | 7348/10833 [01:01<00:27, 127.32it/s] 68%|██████▊ | 7361/10833 [01:01<00:27, 127.60it/s] 68%|██████▊ | 7374/10833 [01:01<00:27, 127.83it/s] 68%|██████▊ | 7387/10833 [01:02<00:26, 128.03it/s] 68%|██████▊ | 7400/10833 [01:02<00:26, 128.06it/s] 68%|██████▊ | 7413/10833 [01:02<00:26, 128.16it/s] 69%|██████▊ | 7426/10833 [01:02<00:26, 128.11it/s] 69%|██████▊ | 7439/10833 [01:02<00:26, 128.09it/s] 69%|██████▉ | 7452/10833 [01:02<00:26, 128.19it/s] 69%|██████▉ | 7465/10833 [01:02<00:26, 128.01it/s] 69%|██████▉ | 7478/10833 [01:02<00:26, 127.99it/s] 69%|██████▉ | 7491/10833 [01:02<00:26, 127.18it/s] 69%|██████▉ | 7504/10833 [01:02<00:26, 127.43it/s] 69%|██████▉ | 7517/10833 [01:03<00:26, 127.39it/s] 70%|██████▉ | 7530/10833 [01:03<00:25, 127.64it/s] 70%|██████▉ | 7543/10833 [01:03<00:25, 127.84it/s] 70%|██████▉ | 7556/10833 [01:03<00:25, 127.87it/s] 70%|██████▉ | 7569/10833 [01:03<00:25, 127.84it/s] 70%|██████▉ | 7582/10833 [01:03<00:25, 127.93it/s] 70%|███████ | 7595/10833 [01:03<00:25, 128.00it/s] 70%|███████ | 7608/10833 [01:03<00:25, 128.06it/s] 70%|███████ | 7621/10833 [01:03<00:25, 128.10it/s] 70%|███████ | 7634/10833 [01:03<00:24, 128.14it/s] 71%|███████ | 7647/10833 [01:04<00:24, 128.12it/s] 71%|███████ | 7660/10833 [01:04<00:24, 128.14it/s] 71%|███████ | 7673/10833 [01:04<00:24, 127.23it/s] 71%|███████ | 7686/10833 [01:04<00:25, 123.92it/s] 71%|███████ | 7699/10833 [01:04<00:25, 125.15it/s] 71%|███████ | 7712/10833 [01:04<00:24, 125.96it/s] 71%|███████▏ | 7725/10833 [01:04<00:24, 126.55it/s] 71%|███████▏ | 7738/10833 [01:04<00:24, 127.10it/s] 72%|███████▏ | 7751/10833 [01:04<00:24, 127.33it/s] 72%|███████▏ | 7764/10833 [01:04<00:24, 127.58it/s] 72%|███████▏ | 7777/10833 [01:05<00:23, 127.71it/s] 72%|███████▏ | 7790/10833 [01:05<00:23, 127.89it/s] 72%|███████▏ | 7803/10833 [01:05<00:23, 127.92it/s] 72%|███████▏ | 7816/10833 [01:05<00:23, 127.93it/s] 72%|███████▏ | 7829/10833 [01:05<00:23, 128.03it/s] 72%|███████▏ | 7842/10833 [01:05<00:23, 127.23it/s] 73%|███████▎ | 7855/10833 [01:05<00:23, 127.13it/s] 73%|███████▎ | 7868/10833 [01:05<00:23, 127.47it/s] 73%|███████▎ | 7881/10833 [01:05<00:23, 127.61it/s] 73%|███████▎ | 7894/10833 [01:05<00:23, 127.73it/s] 73%|███████▎ | 7907/10833 [01:06<00:22, 127.91it/s] 73%|███████▎ | 7920/10833 [01:06<00:22, 127.88it/s] 73%|███████▎ | 7933/10833 [01:06<00:22, 128.04it/s] 73%|███████▎ | 7946/10833 [01:06<00:22, 128.06it/s] 73%|███████▎ | 7959/10833 [01:06<00:22, 128.04it/s] 74%|███████▎ | 7972/10833 [01:06<00:22, 128.03it/s] 74%|███████▎ | 7985/10833 [01:06<00:22, 128.14it/s] 74%|███████▍ | 7998/10833 [01:06<00:22, 128.14it/s] 74%|███████▍ | 8011/10833 [01:06<00:22, 127.23it/s] 74%|███████▍ | 8024/10833 [01:07<00:22, 127.19it/s] 74%|███████▍ | 8037/10833 [01:07<00:21, 127.37it/s] 74%|███████▍ | 8050/10833 [01:07<00:21, 127.59it/s] 74%|███████▍ | 8063/10833 [01:07<00:21, 127.62it/s] 75%|███████▍ | 8076/10833 [01:07<00:21, 127.80it/s] 75%|███████▍ | 8089/10833 [01:07<00:21, 127.88it/s] 75%|███████▍ | 8102/10833 [01:07<00:21, 127.99it/s] 75%|███████▍ | 8115/10833 [01:07<00:21, 128.07it/s] 75%|███████▌ | 8128/10833 [01:07<00:21, 128.15it/s] 75%|███████▌ | 8141/10833 [01:07<00:21, 128.18it/s] 75%|███████▌ | 8154/10833 [01:08<00:20, 128.10it/s] 75%|███████▌ | 8167/10833 [01:08<00:20, 128.12it/s] 76%|███████▌ | 8180/10833 [01:08<00:20, 127.25it/s] 76%|███████▌ | 8193/10833 [01:08<00:20, 127.55it/s] 76%|███████▌ | 8206/10833 [01:08<00:20, 127.40it/s] 76%|███████▌ | 8219/10833 [01:08<00:20, 127.66it/s] 76%|███████▌ | 8232/10833 [01:08<00:20, 127.82it/s] 76%|███████▌ | 8245/10833 [01:08<00:20, 127.96it/s] 76%|███████▌ | 8258/10833 [01:08<00:20, 128.10it/s] 76%|███████▋ | 8271/10833 [01:08<00:20, 128.10it/s] 76%|███████▋ | 8284/10833 [01:09<00:19, 128.19it/s] 77%|███████▋ | 8297/10833 [01:09<00:19, 128.11it/s] 77%|███████▋ | 8310/10833 [01:09<00:19, 128.11it/s] 77%|███████▋ | 8323/10833 [01:09<00:19, 128.09it/s] 77%|███████▋ | 8336/10833 [01:09<00:19, 128.11it/s] 77%|███████▋ | 8349/10833 [01:09<00:19, 127.36it/s] 77%|███████▋ | 8362/10833 [01:09<00:19, 127.60it/s] 77%|███████▋ | 8375/10833 [01:09<00:19, 127.42it/s] 77%|███████▋ | 8388/10833 [01:09<00:19, 127.64it/s] 78%|███████▊ | 8401/10833 [01:09<00:19, 127.73it/s] 78%|███████▊ | 8414/10833 [01:10<00:18, 127.75it/s] 78%|███████▊ | 8427/10833 [01:10<00:18, 127.97it/s] 78%|███████▊ | 8440/10833 [01:10<00:18, 128.00it/s] 78%|███████▊ | 8453/10833 [01:10<00:18, 127.88it/s] 78%|███████▊ | 8466/10833 [01:10<00:18, 127.80it/s] 78%|███████▊ | 8479/10833 [01:10<00:18, 127.92it/s] 78%|███████▊ | 8492/10833 [01:10<00:18, 128.05it/s] 79%|███████▊ | 8505/10833 [01:10<00:18, 128.01it/s] 79%|███████▊ | 8518/10833 [01:10<00:18, 127.30it/s] 79%|███████▉ | 8531/10833 [01:10<00:18, 127.57it/s] 79%|███████▉ | 8544/10833 [01:11<00:17, 127.42it/s] 79%|███████▉ | 8557/10833 [01:11<00:17, 127.54it/s] 79%|███████▉ | 8570/10833 [01:11<00:17, 127.66it/s] 79%|███████▉ | 8583/10833 [01:11<00:17, 127.76it/s] 79%|███████▉ | 8596/10833 [01:11<00:17, 127.90it/s] 79%|███████▉ | 8609/10833 [01:11<00:17, 128.02it/s] 80%|███████▉ | 8622/10833 [01:11<00:17, 128.07it/s] 80%|███████▉ | 8635/10833 [01:11<00:17, 128.09it/s] 80%|███████▉ | 8648/10833 [01:11<00:17, 128.10it/s] 80%|███████▉ | 8661/10833 [01:11<00:16, 128.07it/s] 80%|████████ | 8674/10833 [01:12<00:16, 128.04it/s] 80%|████████ | 8687/10833 [01:12<00:16, 128.13it/s] 80%|████████ | 8700/10833 [01:12<00:16, 127.24it/s] 80%|████████ | 8713/10833 [01:12<00:16, 127.53it/s] 81%|████████ | 8726/10833 [01:12<00:16, 127.33it/s] 81%|████████ | 8739/10833 [01:12<00:16, 127.53it/s] 81%|████████ | 8752/10833 [01:12<00:16, 127.74it/s] 81%|████████ | 8765/10833 [01:12<00:16, 127.80it/s] 81%|████████ | 8778/10833 [01:12<00:16, 127.87it/s] 81%|████████ | 8791/10833 [01:12<00:15, 127.94it/s] 81%|████████▏ | 8804/10833 [01:13<00:15, 127.92it/s] 81%|████████▏ | 8817/10833 [01:13<00:15, 127.95it/s] 82%|████████▏ | 8830/10833 [01:13<00:15, 128.08it/s] 82%|████████▏ | 8843/10833 [01:13<00:15, 128.02it/s] 82%|████████▏ | 8856/10833 [01:13<00:15, 128.08it/s] 82%|████████▏ | 8869/10833 [01:13<00:15, 127.19it/s] 82%|████████▏ | 8882/10833 [01:13<00:15, 127.48it/s] 82%|████████▏ | 8895/10833 [01:13<00:15, 127.38it/s] 82%|████████▏ | 8908/10833 [01:13<00:15, 127.59it/s] 82%|████████▏ | 8921/10833 [01:14<00:14, 127.73it/s] 82%|████████▏ | 8934/10833 [01:14<00:14, 127.84it/s] 83%|████████▎ | 8947/10833 [01:14<00:14, 128.05it/s] 83%|████████▎ | 8960/10833 [01:14<00:14, 127.99it/s] 83%|████████▎ | 8973/10833 [01:14<00:14, 128.11it/s] 83%|████████▎ | 8986/10833 [01:14<00:14, 128.13it/s] 83%|████████▎ | 8999/10833 [01:14<00:14, 128.11it/s] 83%|████████▎ | 9012/10833 [01:14<00:14, 128.14it/s] 83%|████████▎ | 9025/10833 [01:14<00:14, 128.15it/s] 83%|████████▎ | 9038/10833 [01:14<00:14, 127.24it/s] 84%|████████▎ | 9051/10833 [01:15<00:13, 127.57it/s] 84%|████████▎ | 9064/10833 [01:15<00:13, 127.45it/s] 84%|████████▍ | 9077/10833 [01:15<00:13, 127.71it/s] 84%|████████▍ | 9090/10833 [01:15<00:13, 127.85it/s] 84%|████████▍ | 9103/10833 [01:15<00:13, 128.06it/s] 84%|████████▍ | 9116/10833 [01:15<00:13, 128.06it/s] 84%|████████▍ | 9129/10833 [01:15<00:13, 128.08it/s] 84%|████████▍ | 9142/10833 [01:15<00:13, 128.11it/s] 85%|████████▍ | 9155/10833 [01:15<00:13, 128.16it/s] 85%|████████▍ | 9168/10833 [01:15<00:13, 128.00it/s] 85%|████████▍ | 9181/10833 [01:16<00:12, 128.05it/s] 85%|████████▍ | 9194/10833 [01:16<00:12, 128.01it/s] 85%|████████▍ | 9207/10833 [01:16<00:12, 127.22it/s] 85%|████████▌ | 9220/10833 [01:16<00:12, 127.41it/s] 85%|████████▌ | 9233/10833 [01:16<00:12, 127.64it/s] 85%|████████▌ | 9246/10833 [01:16<00:12, 127.41it/s] 85%|████████▌ | 9259/10833 [01:16<00:12, 127.59it/s] 86%|████████▌ | 9272/10833 [01:16<00:12, 127.84it/s] 86%|████████▌ | 9285/10833 [01:16<00:12, 127.93it/s] 86%|████████▌ | 9298/10833 [01:16<00:11, 128.10it/s] 86%|████████▌ | 9311/10833 [01:17<00:11, 128.11it/s] 86%|████████▌ | 9324/10833 [01:17<00:11, 128.14it/s] 86%|████████▌ | 9337/10833 [01:17<00:11, 128.07it/s] 86%|████████▋ | 9350/10833 [01:17<00:11, 128.02it/s] 86%|████████▋ | 9363/10833 [01:17<00:11, 128.01it/s] 87%|████████▋ | 9376/10833 [01:17<00:11, 127.18it/s] 87%|████████▋ | 9389/10833 [01:17<00:11, 127.50it/s] 87%|████████▋ | 9402/10833 [01:17<00:11, 127.79it/s] 87%|████████▋ | 9415/10833 [01:17<00:11, 127.58it/s] 87%|████████▋ | 9428/10833 [01:17<00:10, 127.80it/s] 87%|████████▋ | 9441/10833 [01:18<00:10, 127.86it/s] 87%|████████▋ | 9454/10833 [01:18<00:10, 127.78it/s] 87%|████████▋ | 9467/10833 [01:18<00:10, 127.85it/s] 88%|████████▊ | 9480/10833 [01:18<00:10, 127.95it/s] 88%|████████▊ | 9493/10833 [01:18<00:10, 128.02it/s] 88%|████████▊ | 9506/10833 [01:18<00:10, 128.10it/s] 88%|████████▊ | 9519/10833 [01:18<00:10, 128.09it/s] 88%|████████▊ | 9532/10833 [01:18<00:10, 128.15it/s] 88%|████████▊ | 9545/10833 [01:18<00:10, 128.14it/s] 88%|████████▊ | 9558/10833 [01:19<00:10, 127.21it/s] 88%|████████▊ | 9571/10833 [01:19<00:09, 127.47it/s] 88%|████████▊ | 9584/10833 [01:19<00:09, 127.31it/s] 89%|████████▊ | 9597/10833 [01:19<00:09, 127.61it/s] 89%|████████▊ | 9610/10833 [01:19<00:09, 127.76it/s] 89%|████████▉ | 9623/10833 [01:19<00:09, 127.94it/s] 89%|████████▉ | 9636/10833 [01:19<00:09, 128.01it/s] 89%|████████▉ | 9649/10833 [01:19<00:09, 128.11it/s] 89%|████████▉ | 9662/10833 [01:19<00:09, 128.18it/s] 89%|████████▉ | 9675/10833 [01:19<00:09, 128.25it/s] 89%|████████▉ | 9688/10833 [01:20<00:08, 128.25it/s] 90%|████████▉ | 9701/10833 [01:20<00:08, 128.31it/s] 90%|████████▉ | 9714/10833 [01:20<00:08, 128.22it/s] 90%|████████▉ | 9727/10833 [01:20<00:08, 127.37it/s] 90%|████████▉ | 9740/10833 [01:20<00:08, 127.62it/s] 90%|█████████ | 9753/10833 [01:20<00:08, 127.76it/s] 90%|█████████ | 9766/10833 [01:20<00:08, 124.43it/s] 90%|█████████ | 9779/10833 [01:20<00:08, 125.49it/s] 90%|█████████ | 9792/10833 [01:20<00:08, 126.23it/s] 91%|█████████ | 9805/10833 [01:20<00:08, 126.78it/s] 91%|█████████ | 9818/10833 [01:21<00:07, 127.11it/s] 91%|█████████ | 9831/10833 [01:21<00:07, 127.38it/s] 91%|█████████ | 9844/10833 [01:21<00:07, 127.54it/s] 91%|█████████ | 9857/10833 [01:21<00:07, 127.70it/s] 91%|█████████ | 9870/10833 [01:21<00:07, 127.79it/s] 91%|█████████ | 9883/10833 [01:21<00:07, 127.86it/s] 91%|█████████▏| 9896/10833 [01:21<00:07, 126.85it/s] 91%|█████████▏| 9909/10833 [01:21<00:07, 127.27it/s] 92%|█████████▏| 9922/10833 [01:21<00:07, 127.53it/s] 92%|█████████▏| 9935/10833 [01:21<00:07, 127.34it/s] 92%|█████████▏| 9948/10833 [01:22<00:06, 127.50it/s] 92%|█████████▏| 9961/10833 [01:22<00:06, 127.79it/s] 92%|█████████▏| 9974/10833 [01:22<00:06, 127.83it/s] 92%|█████████▏| 9987/10833 [01:22<00:06, 127.89it/s] 92%|█████████▏| 10000/10833 [01:22<00:06, 127.90it/s] 92%|█████████▏| 10013/10833 [01:22<00:06, 128.04it/s] 93%|█████████▎| 10026/10833 [01:22<00:06, 128.09it/s] 93%|█████████▎| 10039/10833 [01:22<00:06, 128.02it/s] 93%|█████████▎| 10052/10833 [01:22<00:06, 128.06it/s] 93%|█████████▎| 10065/10833 [01:22<00:06, 127.19it/s] 93%|█████████▎| 10078/10833 [01:23<00:05, 127.45it/s] 93%|█████████▎| 10091/10833 [01:23<00:05, 127.67it/s] 93%|█████████▎| 10104/10833 [01:23<00:05, 127.46it/s] 93%|█████████▎| 10117/10833 [01:23<00:05, 127.54it/s] 94%|█████████▎| 10130/10833 [01:23<00:05, 127.80it/s] 94%|█████████▎| 10143/10833 [01:23<00:05, 127.85it/s] 94%|█████████▍| 10156/10833 [01:23<00:05, 127.92it/s] 94%|█████████▍| 10169/10833 [01:23<00:05, 128.02it/s] 94%|█████████▍| 10182/10833 [01:23<00:05, 128.04it/s] 94%|█████████▍| 10195/10833 [01:23<00:04, 128.10it/s] 94%|█████████▍| 10208/10833 [01:24<00:04, 127.99it/s] 94%|█████████▍| 10221/10833 [01:24<00:04, 128.05it/s] 94%|█████████▍| 10234/10833 [01:24<00:04, 126.95it/s] 95%|█████████▍| 10247/10833 [01:24<00:04, 127.25it/s] 95%|█████████▍| 10260/10833 [01:24<00:04, 127.38it/s] 95%|█████████▍| 10273/10833 [01:24<00:04, 127.15it/s] 95%|█████████▍| 10286/10833 [01:24<00:04, 127.49it/s] 95%|█████████▌| 10299/10833 [01:24<00:04, 127.57it/s] 95%|█████████▌| 10312/10833 [01:24<00:04, 127.70it/s] 95%|█████████▌| 10325/10833 [01:25<00:03, 127.74it/s] 95%|█████████▌| 10338/10833 [01:25<00:03, 127.82it/s] 96%|█████████▌| 10351/10833 [01:25<00:03, 127.95it/s] 96%|█████████▌| 10364/10833 [01:25<00:03, 128.03it/s] 96%|█████████▌| 10377/10833 [01:25<00:03, 128.02it/s] 96%|█████████▌| 10390/10833 [01:25<00:03, 128.05it/s] 96%|█████████▌| 10403/10833 [01:25<00:03, 127.15it/s] 96%|█████████▌| 10416/10833 [01:25<00:03, 127.49it/s] 96%|█████████▋| 10429/10833 [01:25<00:03, 127.76it/s] 96%|█████████▋| 10442/10833 [01:25<00:03, 127.77it/s] 97%|█████████▋| 10455/10833 [01:26<00:02, 127.37it/s] 97%|█████████▋| 10468/10833 [01:26<00:02, 127.53it/s] 97%|█████████▋| 10481/10833 [01:26<00:02, 127.67it/s] 97%|█████████▋| 10494/10833 [01:26<00:02, 127.70it/s] 97%|█████████▋| 10507/10833 [01:26<00:02, 127.76it/s] 97%|█████████▋| 10520/10833 [01:26<00:02, 127.82it/s] 97%|█████████▋| 10533/10833 [01:26<00:02, 127.87it/s] 97%|█████████▋| 10546/10833 [01:26<00:02, 127.70it/s] 97%|█████████▋| 10559/10833 [01:26<00:02, 127.82it/s] 98%|█████████▊| 10572/10833 [01:26<00:02, 127.97it/s] 98%|█████████▊| 10585/10833 [01:27<00:01, 126.99it/s] 98%|█████████▊| 10598/10833 [01:27<00:01, 127.31it/s] 98%|█████████▊| 10611/10833 [01:27<00:01, 127.39it/s] 98%|█████████▊| 10624/10833 [01:27<00:01, 127.17it/s] 98%|█████████▊| 10637/10833 [01:27<00:01, 127.37it/s] 98%|█████████▊| 10650/10833 [01:27<00:01, 127.58it/s] 98%|█████████▊| 10663/10833 [01:27<00:01, 127.80it/s] 99%|█████████▊| 10676/10833 [01:27<00:01, 127.78it/s] 99%|█████████▊| 10689/10833 [01:27<00:01, 127.89it/s] 99%|█████████▉| 10702/10833 [01:27<00:01, 127.91it/s] 99%|█████████▉| 10715/10833 [01:28<00:00, 127.88it/s] 99%|█████████▉| 10728/10833 [01:28<00:00, 127.96it/s] 99%|█████████▉| 10741/10833 [01:28<00:00, 127.84it/s] 99%|█████████▉| 10754/10833 [01:28<00:00, 126.95it/s] 99%|█████████▉| 10767/10833 [01:28<00:00, 127.22it/s] 100%|█████████▉| 10780/10833 [01:28<00:00, 127.42it/s] 100%|█████████▉| 10793/10833 [01:28<00:00, 127.20it/s] 100%|█████████▉| 10806/10833 [01:28<00:00, 127.31it/s] 100%|█████████▉| 10819/10833 [01:28<00:00, 127.45it/s] 100%|█████████▉| 10832/10833 [01:28<00:00, 127.70it/s] 100%|██████████| 10833/10833 [01:28<00:00, 121.73it/s]
INFO:TestAccuracy:Batch size is 1, F1: 85.85987, Exact Match:78.75935
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
INFO:BackendDCU:Batch size is 1, QPS: 135, Avg Latency:7.38, Tail Latency:9.51
INFO:BackendDCU:Batch size is 2, QPS: 176, Avg Latency:11.32, Tail Latency:13.64
INFO:BackendDCU:Batch size is 4, QPS: 203, Avg Latency:19.63, Tail Latency:22.32
INFO:BackendDCU:Batch size is 8, QPS: 234, Avg Latency:34.14, Tail Latency:36.54
INFO:BackendDCU:Batch size is 16, QPS: 255, Avg Latency:62.74, Tail Latency:64.84
INFO:BackendDCU:Batch size is 32, QPS: 266, Avg Latency:119.9, Tail Latency:122.05
INFO:BackendDCU:Batch size is 64, QPS: 270, Avg Latency:236.36, Tail Latency:238.39
INFO:BackendDCU:Batch size is 128, QPS: 272, Avg Latency:469.45, Tail Latency:471.21
INFO:BackendDCU:Batch size is 256, QPS: 268, Avg Latency:953.39, Tail Latency:954.79
INFO:BackendDCU:Batch size is 512, QPS: 265, Avg Latency:1927.56, Tail Latency:1930.66
ByteMLPerf/byte_infer_perf/general_perf/log/bert-torch-fp32.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
INFO:PerfEngine:******************************************* Start to test model: bert-torch-fp32. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_squad
INFO:SQUAD:Initial...
INFO:SQUAD:Preprocessing...
INFO:SQUAD:Rebatching batch size to: 10833 ...
0%| | 0/1 [00:00<?, ?it/s] 100%|██████████| 1/1 [00:00<00:00, 112.89it/s]
INFO:PerfEngine:Start to compile the model...
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:SQUAD:Rebatching batch size to: 1 ...
0%| | 0/10833 [00:00<?, ?it/s] 100%|██████████| 10833/10833 [00:00<00:00, 357683.52it/s]
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/10833 [00:00<?, ?it/s] 0%| | 1/10833 [00:03<11:25:53, 3.80s/it] 0%| | 6/10833 [00:03<1:27:24, 2.06it/s] 0%| | 19/10833 [00:04<21:38, 8.33it/s] 0%| | 32/10833 [00:04<11:06, 16.21it/s] 0%| | 45/10833 [00:04<06:59, 25.74it/s] 1%| | 57/10833 [00:04<05:00, 35.85it/s] 1%| | 70/10833 [00:04<03:44, 47.92it/s] 1%| | 83/10833 [00:04<02:58, 60.23it/s] 1%| | 96/10833 [00:04<02:29, 72.03it/s] 1%| | 109/10833 [00:04<02:09, 82.65it/s] 1%| | 122/10833 [00:04<01:56, 91.67it/s] 1%| | 135/10833 [00:04<01:47, 99.07it/s] 1%|▏ | 148/10833 [00:05<01:41, 104.81it/s] 1%|▏ | 161/10833 [00:05<01:37, 109.15it/s] 2%|▏ | 174/10833 [00:05<01:34, 112.43it/s] 2%|▏ | 187/10833 [00:05<01:32, 114.83it/s] 2%|▏ | 200/10833 [00:05<01:31, 116.51it/s] 2%|▏ | 213/10833 [00:05<01:30, 117.73it/s] 2%|▏ | 226/10833 [00:05<01:30, 117.79it/s] 2%|▏ | 239/10833 [00:05<01:29, 118.54it/s] 2%|▏ | 252/10833 [00:05<01:28, 119.14it/s] 2%|▏ | 265/10833 [00:06<01:28, 119.53it/s] 3%|▎ | 278/10833 [00:06<01:28, 119.80it/s] 3%|▎ | 291/10833 [00:06<01:27, 120.04it/s] 3%|▎ | 304/10833 [00:06<01:27, 120.12it/s] 3%|▎ | 317/10833 [00:06<01:27, 120.29it/s] 3%|▎ | 330/10833 [00:06<01:27, 120.48it/s] 3%|▎ | 343/10833 [00:06<01:26, 120.66it/s] 3%|▎ | 356/10833 [00:06<01:26, 120.74it/s] 3%|▎ | 369/10833 [00:06<01:26, 120.71it/s] 4%|▎ | 382/10833 [00:07<01:27, 119.90it/s] 4%|▎ | 395/10833 [00:07<01:26, 120.19it/s] 4%|▍ | 408/10833 [00:07<01:26, 120.29it/s] 4%|▍ | 421/10833 [00:07<01:26, 120.22it/s] 4%|▍ | 434/10833 [00:07<01:26, 120.24it/s] 4%|▍ | 447/10833 [00:07<01:26, 120.45it/s] 4%|▍ | 460/10833 [00:07<01:26, 120.40it/s] 4%|▍ | 473/10833 [00:07<01:26, 120.45it/s] 4%|▍ | 486/10833 [00:07<01:25, 120.49it/s] 5%|▍ | 499/10833 [00:07<01:25, 120.49it/s] 5%|▍ | 512/10833 [00:08<01:25, 120.50it/s] 5%|▍ | 525/10833 [00:08<01:25, 120.62it/s] 5%|▍ | 538/10833 [00:08<01:25, 119.86it/s] 5%|▌ | 551/10833 [00:08<01:25, 120.08it/s] 5%|▌ | 564/10833 [00:08<01:25, 120.25it/s] 5%|▌ | 577/10833 [00:08<01:25, 120.25it/s] 5%|▌ | 590/10833 [00:08<01:25, 120.29it/s] 6%|▌ | 603/10833 [00:08<01:24, 120.39it/s] 6%|▌ | 616/10833 [00:08<01:24, 120.38it/s] 6%|▌ | 629/10833 [00:09<01:24, 120.49it/s] 6%|▌ | 642/10833 [00:09<01:24, 120.52it/s] 6%|▌ | 655/10833 [00:09<01:24, 120.59it/s] 6%|▌ | 668/10833 [00:09<01:24, 120.46it/s] 6%|▋ | 681/10833 [00:09<01:24, 120.46it/s] 6%|▋ | 694/10833 [00:09<01:24, 120.48it/s] 7%|▋ | 707/10833 [00:09<01:24, 119.85it/s] 7%|▋ | 720/10833 [00:09<01:24, 120.12it/s] 7%|▋ | 733/10833 [00:09<01:23, 120.27it/s] 7%|▋ | 746/10833 [00:10<01:23, 120.26it/s] 7%|▋ | 759/10833 [00:10<01:23, 120.29it/s] 7%|▋ | 772/10833 [00:10<01:23, 120.37it/s] 7%|▋ | 785/10833 [00:10<01:23, 120.29it/s] 7%|▋ | 798/10833 [00:10<01:23, 120.32it/s] 7%|▋ | 811/10833 [00:10<01:23, 120.33it/s] 8%|▊ | 824/10833 [00:10<01:23, 120.25it/s] 8%|▊ | 837/10833 [00:10<01:23, 120.26it/s] 8%|▊ | 850/10833 [00:10<01:23, 120.27it/s] 8%|▊ | 863/10833 [00:11<01:23, 119.67it/s] 8%|▊ | 876/10833 [00:11<01:22, 119.97it/s] 8%|▊ | 889/10833 [00:11<01:22, 120.02it/s] 8%|▊ | 902/10833 [00:11<01:22, 120.07it/s] 8%|▊ | 915/10833 [00:11<01:22, 120.16it/s] 9%|▊ | 928/10833 [00:11<01:22, 120.07it/s] 9%|▊ | 941/10833 [00:11<01:22, 120.17it/s] 9%|▉ | 954/10833 [00:11<01:22, 120.36it/s] 9%|▉ | 967/10833 [00:11<01:21, 120.46it/s] 9%|▉ | 980/10833 [00:11<01:21, 120.58it/s] 9%|▉ | 993/10833 [00:12<01:21, 120.47it/s] 9%|▉ | 1006/10833 [00:12<01:21, 120.32it/s] 9%|▉ | 1019/10833 [00:12<01:21, 120.36it/s] 10%|▉ | 1032/10833 [00:12<01:21, 119.69it/s] 10%|▉ | 1045/10833 [00:12<01:21, 119.90it/s] 10%|▉ | 1057/10833 [00:12<02:44, 59.29it/s] 10%|▉ | 1070/10833 [00:13<02:19, 70.22it/s] 10%|▉ | 1083/10833 [00:13<02:01, 80.42it/s] 10%|█ | 1096/10833 [00:13<01:48, 89.37it/s] 10%|█ | 1109/10833 [00:13<01:40, 96.97it/s] 10%|█ | 1122/10833 [00:13<01:34, 103.05it/s] 10%|█ | 1135/10833 [00:13<01:29, 107.77it/s] 11%|█ | 1147/10833 [00:13<01:27, 110.48it/s] 11%|█ | 1160/10833 [00:13<01:25, 113.41it/s] 11%|█ | 1173/10833 [00:13<01:23, 115.49it/s] 11%|█ | 1186/10833 [00:14<01:22, 117.09it/s] 11%|█ | 1199/10833 [00:14<01:21, 118.13it/s] 11%|█ | 1211/10833 [00:14<01:21, 118.63it/s] 11%|█▏ | 1224/10833 [00:14<01:20, 119.25it/s] 11%|█▏ | 1237/10833 [00:14<01:20, 119.67it/s] 12%|█▏ | 1250/10833 [00:14<01:19, 119.97it/s] 12%|█▏ | 1263/10833 [00:14<01:19, 120.22it/s] 12%|█▏ | 1276/10833 [00:14<01:19, 120.37it/s] 12%|█▏ | 1289/10833 [00:14<01:19, 120.45it/s] 12%|█▏ | 1302/10833 [00:15<01:19, 119.88it/s] 12%|█▏ | 1315/10833 [00:15<01:19, 120.17it/s] 12%|█▏ | 1328/10833 [00:15<01:18, 120.38it/s] 12%|█▏ | 1341/10833 [00:15<01:18, 120.47it/s] 12%|█▏ | 1354/10833 [00:15<01:18, 120.61it/s] 13%|█▎ | 1367/10833 [00:15<01:18, 120.65it/s] 13%|█▎ | 1380/10833 [00:15<01:18, 120.63it/s] 13%|█▎ | 1393/10833 [00:15<01:18, 120.69it/s] 13%|█▎ | 1406/10833 [00:15<01:18, 120.73it/s] 13%|█▎ | 1419/10833 [00:15<01:18, 120.69it/s] 13%|█▎ | 1432/10833 [00:16<01:17, 120.64it/s] 13%|█▎ | 1445/10833 [00:16<01:17, 120.67it/s] 13%|█▎ | 1458/10833 [00:16<01:17, 120.69it/s] 14%|█▎ | 1471/10833 [00:16<01:17, 120.05it/s] 14%|█▎ | 1484/10833 [00:16<01:17, 120.29it/s] 14%|█▍ | 1497/10833 [00:16<01:17, 120.46it/s] 14%|█▍ | 1510/10833 [00:16<01:17, 120.53it/s] 14%|█▍ | 1523/10833 [00:16<01:17, 120.66it/s] 14%|█▍ | 1536/10833 [00:16<01:17, 120.67it/s] 14%|█▍ | 1549/10833 [00:17<01:16, 120.74it/s] 14%|█▍ | 1562/10833 [00:17<01:16, 120.73it/s] 15%|█▍ | 1575/10833 [00:17<01:16, 120.75it/s] 15%|█▍ | 1588/10833 [00:17<01:16, 120.79it/s] 15%|█▍ | 1601/10833 [00:17<01:16, 120.73it/s] 15%|█▍ | 1614/10833 [00:17<01:16, 120.70it/s] 15%|█▌ | 1627/10833 [00:17<01:16, 119.97it/s] 15%|█▌ | 1640/10833 [00:17<01:16, 120.25it/s] 15%|█▌ | 1653/10833 [00:17<01:16, 120.36it/s] 15%|█▌ | 1666/10833 [00:18<01:16, 120.45it/s] 15%|█▌ | 1679/10833 [00:18<01:15, 120.54it/s] 16%|█▌ | 1692/10833 [00:18<01:15, 120.61it/s] 16%|█▌ | 1705/10833 [00:18<01:15, 120.59it/s] 16%|█▌ | 1718/10833 [00:18<01:15, 120.36it/s] 16%|█▌ | 1731/10833 [00:18<01:15, 120.29it/s] 16%|█▌ | 1744/10833 [00:18<01:15, 120.36it/s] 16%|█▌ | 1757/10833 [00:18<01:15, 120.37it/s] 16%|█▋ | 1770/10833 [00:18<01:15, 120.36it/s] 16%|█▋ | 1783/10833 [00:19<01:15, 120.39it/s] 17%|█▋ | 1796/10833 [00:19<01:15, 119.72it/s] 17%|█▋ | 1809/10833 [00:19<01:15, 119.89it/s] 17%|█▋ | 1822/10833 [00:19<01:15, 120.10it/s] 17%|█▋ | 1835/10833 [00:19<01:14, 120.24it/s] 17%|█▋ | 1848/10833 [00:19<01:14, 120.35it/s] 17%|█▋ | 1861/10833 [00:19<01:14, 120.38it/s] 17%|█▋ | 1874/10833 [00:19<01:14, 120.41it/s] 17%|█▋ | 1887/10833 [00:19<01:14, 120.37it/s] 18%|█▊ | 1900/10833 [00:19<01:14, 120.44it/s] 18%|█▊ | 1913/10833 [00:20<01:14, 120.47it/s] 18%|█▊ | 1926/10833 [00:20<01:13, 120.59it/s] 18%|█▊ | 1939/10833 [00:20<01:13, 120.58it/s] 18%|█▊ | 1952/10833 [00:20<01:14, 119.92it/s] 18%|█▊ | 1965/10833 [00:20<01:13, 120.06it/s] 18%|█▊ | 1978/10833 [00:20<01:13, 120.19it/s] 18%|█▊ | 1991/10833 [00:20<01:13, 120.30it/s] 18%|█▊ | 2004/10833 [00:20<01:13, 120.39it/s] 19%|█▊ | 2017/10833 [00:20<01:13, 120.42it/s] 19%|█▊ | 2030/10833 [00:21<01:13, 120.44it/s] 19%|█▉ | 2043/10833 [00:21<01:13, 120.29it/s] 19%|█▉ | 2056/10833 [00:21<01:12, 120.44it/s] 19%|█▉ | 2069/10833 [00:21<01:12, 120.43it/s] 19%|█▉ | 2082/10833 [00:21<01:12, 120.40it/s] 19%|█▉ | 2095/10833 [00:21<01:12, 120.48it/s] 19%|█▉ | 2108/10833 [00:21<01:12, 120.45it/s] 20%|█▉ | 2121/10833 [00:21<01:12, 119.82it/s] 20%|█▉ | 2134/10833 [00:21<01:12, 119.95it/s] 20%|█▉ | 2147/10833 [00:22<01:12, 120.08it/s] 20%|█▉ | 2160/10833 [00:22<01:12, 120.08it/s] 20%|██ | 2173/10833 [00:22<01:12, 120.15it/s] 20%|██ | 2186/10833 [00:22<01:11, 120.20it/s] 20%|██ | 2199/10833 [00:22<01:11, 120.23it/s] 20%|██ | 2212/10833 [00:22<01:11, 120.17it/s] 21%|██ | 2225/10833 [00:22<01:11, 120.17it/s] 21%|██ | 2238/10833 [00:22<01:11, 120.22it/s] 21%|██ | 2251/10833 [00:22<01:11, 120.35it/s] 21%|██ | 2264/10833 [00:23<01:11, 120.34it/s] 21%|██ | 2277/10833 [00:23<01:11, 119.58it/s] 21%|██ | 2290/10833 [00:23<01:11, 119.77it/s] 21%|██▏ | 2303/10833 [00:23<01:11, 119.94it/s] 21%|██▏ | 2316/10833 [00:23<01:10, 120.15it/s] 21%|██▏ | 2329/10833 [00:23<01:10, 120.09it/s] 22%|██▏ | 2342/10833 [00:23<01:10, 120.19it/s] 22%|██▏ | 2355/10833 [00:23<01:10, 120.32it/s] 22%|██▏ | 2368/10833 [00:23<01:10, 120.29it/s] 22%|██▏ | 2381/10833 [00:23<01:10, 120.36it/s] 22%|██▏ | 2394/10833 [00:24<01:10, 120.47it/s] 22%|██▏ | 2407/10833 [00:24<01:10, 120.31it/s] 22%|██▏ | 2420/10833 [00:24<01:09, 120.21it/s] 22%|██▏ | 2433/10833 [00:24<01:10, 119.58it/s] 23%|██▎ | 2446/10833 [00:24<01:10, 119.81it/s] 23%|██▎ | 2459/10833 [00:24<01:09, 119.94it/s] 23%|██▎ | 2472/10833 [00:24<01:09, 120.04it/s] 23%|██▎ | 2485/10833 [00:24<01:09, 120.11it/s] 23%|██▎ | 2498/10833 [00:24<01:09, 120.21it/s] 23%|██▎ | 2511/10833 [00:25<01:09, 120.29it/s] 23%|██▎ | 2524/10833 [00:25<01:09, 120.25it/s] 23%|██▎ | 2537/10833 [00:25<01:09, 120.19it/s] 24%|██▎ | 2550/10833 [00:25<01:08, 120.27it/s] 24%|██▎ | 2563/10833 [00:25<01:08, 120.31it/s] 24%|██▍ | 2576/10833 [00:25<01:08, 120.33it/s] 24%|██▍ | 2589/10833 [00:25<01:08, 120.31it/s] 24%|██▍ | 2602/10833 [00:25<01:08, 119.52it/s] 24%|██▍ | 2615/10833 [00:25<01:08, 119.73it/s] 24%|██▍ | 2628/10833 [00:26<01:08, 119.90it/s] 24%|██▍ | 2641/10833 [00:26<01:08, 119.93it/s] 24%|██▍ | 2654/10833 [00:26<01:08, 120.10it/s] 25%|██▍ | 2667/10833 [00:26<01:08, 120.04it/s] 25%|██▍ | 2680/10833 [00:26<01:07, 120.15it/s] 25%|██▍ | 2693/10833 [00:26<01:07, 120.29it/s] 25%|██▍ | 2706/10833 [00:26<01:07, 120.28it/s] 25%|██▌ | 2719/10833 [00:26<01:07, 120.24it/s] 25%|██▌ | 2732/10833 [00:26<01:07, 120.25it/s] 25%|██▌ | 2745/10833 [00:27<01:07, 120.31it/s] 25%|██▌ | 2758/10833 [00:27<01:07, 119.59it/s] 26%|██▌ | 2771/10833 [00:27<01:07, 119.83it/s] 26%|██▌ | 2784/10833 [00:27<01:07, 120.01it/s] 26%|██▌ | 2797/10833 [00:27<01:06, 120.23it/s] 26%|██▌ | 2810/10833 [00:27<01:06, 120.36it/s] 26%|██▌ | 2823/10833 [00:27<01:06, 120.45it/s] 26%|██▌ | 2836/10833 [00:27<01:06, 120.55it/s] 26%|██▋ | 2849/10833 [00:27<01:06, 120.59it/s] 26%|██▋ | 2862/10833 [00:27<01:06, 120.45it/s] 27%|██▋ | 2875/10833 [00:28<01:06, 120.46it/s] 27%|██▋ | 2888/10833 [00:28<01:05, 120.40it/s] 27%|██▋ | 2901/10833 [00:28<01:05, 120.36it/s] 27%|██▋ | 2914/10833 [00:28<01:05, 120.39it/s] 27%|██▋ | 2927/10833 [00:28<01:06, 119.58it/s] 27%|██▋ | 2940/10833 [00:28<01:05, 119.80it/s] 27%|██▋ | 2953/10833 [00:28<01:05, 119.97it/s] 27%|██▋ | 2966/10833 [00:28<01:05, 120.05it/s] 27%|██▋ | 2979/10833 [00:28<01:05, 120.12it/s] 28%|██▊ | 2992/10833 [00:29<01:05, 120.15it/s] 28%|██▊ | 3005/10833 [00:29<01:05, 120.10it/s] 28%|██▊ | 3018/10833 [00:29<01:05, 120.06it/s] 28%|██▊ | 3031/10833 [00:29<01:04, 120.14it/s] 28%|██▊ | 3044/10833 [00:29<01:04, 120.30it/s] 28%|██▊ | 3057/10833 [00:29<01:04, 120.27it/s] 28%|██▊ | 3070/10833 [00:29<01:04, 120.30it/s] 28%|██▊ | 3083/10833 [00:29<01:04, 119.55it/s] 29%|██▊ | 3096/10833 [00:29<01:04, 119.78it/s] 29%|██▊ | 3108/10833 [00:30<01:04, 119.75it/s] 29%|██▉ | 3120/10833 [00:30<01:05, 117.39it/s] 29%|██▉ | 3133/10833 [00:30<01:05, 118.33it/s] 29%|██▉ | 3146/10833 [00:30<01:04, 118.89it/s] 29%|██▉ | 3159/10833 [00:30<01:04, 119.43it/s] 29%|██▉ | 3172/10833 [00:30<01:03, 119.71it/s] 29%|██▉ | 3184/10833 [00:30<01:03, 119.75it/s] 30%|██▉ | 3197/10833 [00:30<01:03, 119.90it/s] 30%|██▉ | 3210/10833 [00:30<01:03, 119.99it/s] 30%|██▉ | 3223/10833 [00:31<01:03, 120.05it/s] 30%|██▉ | 3236/10833 [00:31<01:03, 119.43it/s] 30%|██▉ | 3249/10833 [00:31<01:03, 119.65it/s] 30%|███ | 3262/10833 [00:31<01:03, 119.81it/s] 30%|███ | 3275/10833 [00:31<01:03, 119.94it/s] 30%|███ | 3287/10833 [00:31<01:02, 119.92it/s] 30%|███ | 3300/10833 [00:31<01:02, 120.02it/s] 31%|███ | 3313/10833 [00:31<01:02, 120.07it/s] 31%|███ | 3326/10833 [00:31<01:02, 120.11it/s] 31%|███ | 3339/10833 [00:31<01:02, 120.10it/s] 31%|███ | 3352/10833 [00:32<01:02, 120.10it/s] 31%|███ | 3365/10833 [00:32<01:02, 120.15it/s] 31%|███ | 3378/10833 [00:32<01:02, 120.20it/s] 31%|███▏ | 3391/10833 [00:32<01:01, 120.18it/s] 31%|███▏ | 3404/10833 [00:32<01:02, 119.43it/s] 32%|███▏ | 3417/10833 [00:32<01:01, 119.71it/s] 32%|███▏ | 3430/10833 [00:32<01:01, 119.87it/s] 32%|███▏ | 3443/10833 [00:32<01:01, 119.96it/s] 32%|███▏ | 3456/10833 [00:32<01:01, 120.02it/s] 32%|███▏ | 3469/10833 [00:33<01:01, 120.02it/s] 32%|███▏ | 3482/10833 [00:33<01:01, 120.15it/s] 32%|███▏ | 3495/10833 [00:33<01:01, 120.25it/s] 32%|███▏ | 3508/10833 [00:33<01:00, 120.22it/s] 33%|███▎ | 3521/10833 [00:33<01:00, 120.19it/s] 33%|███▎ | 3534/10833 [00:33<01:00, 120.20it/s] 33%|███▎ | 3547/10833 [00:33<01:00, 120.18it/s] 33%|███▎ | 3560/10833 [00:33<01:00, 119.51it/s] 33%|███▎ | 3573/10833 [00:33<01:00, 119.78it/s] 33%|███▎ | 3586/10833 [00:34<01:00, 119.98it/s] 33%|███▎ | 3599/10833 [00:34<01:00, 120.04it/s] 33%|███▎ | 3612/10833 [00:34<01:00, 120.11it/s] 33%|███▎ | 3625/10833 [00:34<01:00, 120.03it/s] 34%|███▎ | 3638/10833 [00:34<00:59, 120.00it/s] 34%|███▎ | 3651/10833 [00:34<00:59, 120.05it/s] 34%|███▍ | 3664/10833 [00:34<00:59, 120.08it/s] 34%|███▍ | 3677/10833 [00:34<00:59, 120.15it/s] 34%|███▍ | 3690/10833 [00:34<00:59, 120.22it/s] 34%|███▍ | 3703/10833 [00:35<00:59, 120.18it/s] 34%|███▍ | 3716/10833 [00:35<00:59, 120.24it/s] 34%|███▍ | 3729/10833 [00:35<00:59, 119.60it/s] 35%|███▍ | 3742/10833 [00:35<00:59, 119.75it/s] 35%|███▍ | 3755/10833 [00:35<00:59, 119.85it/s] 35%|███▍ | 3768/10833 [00:35<00:58, 119.99it/s] 35%|███▍ | 3781/10833 [00:35<00:58, 120.09it/s] 35%|███▌ | 3794/10833 [00:35<00:58, 120.15it/s] 35%|███▌ | 3807/10833 [00:35<00:58, 120.11it/s] 35%|███▌ | 3820/10833 [00:35<00:58, 120.21it/s] 35%|███▌ | 3833/10833 [00:36<00:58, 120.17it/s] 36%|███▌ | 3846/10833 [00:36<00:58, 120.09it/s] 36%|███▌ | 3859/10833 [00:36<00:58, 120.04it/s] 36%|███▌ | 3872/10833 [00:36<00:57, 120.20it/s] 36%|███▌ | 3885/10833 [00:36<00:58, 119.48it/s] 36%|███▌ | 3898/10833 [00:36<00:57, 119.78it/s] 36%|███▌ | 3910/10833 [00:36<00:57, 119.82it/s] 36%|███▌ | 3922/10833 [00:36<00:57, 119.83it/s] 36%|███▋ | 3935/10833 [00:36<00:57, 119.92it/s] 36%|███▋ | 3948/10833 [00:37<00:57, 120.04it/s] 37%|███▋ | 3961/10833 [00:37<00:57, 120.10it/s] 37%|███▋ | 3974/10833 [00:37<00:57, 120.15it/s] 37%|███▋ | 3987/10833 [00:37<00:56, 120.12it/s] 37%|███▋ | 4000/10833 [00:37<00:56, 120.24it/s] 37%|███▋ | 4013/10833 [00:37<00:56, 120.20it/s] 37%|███▋ | 4026/10833 [00:37<00:56, 120.25it/s] 37%|███▋ | 4039/10833 [00:37<00:56, 120.21it/s] 37%|███▋ | 4052/10833 [00:37<00:56, 119.49it/s] 38%|███▊ | 4065/10833 [00:38<00:56, 119.66it/s] 38%|███▊ | 4078/10833 [00:38<00:56, 119.85it/s] 38%|███▊ | 4090/10833 [00:38<00:56, 119.75it/s] 38%|███▊ | 4103/10833 [00:38<00:56, 119.97it/s] 38%|███▊ | 4116/10833 [00:38<00:55, 120.08it/s] 38%|███▊ | 4129/10833 [00:38<00:55, 120.09it/s] 38%|███▊ | 4142/10833 [00:38<00:55, 119.97it/s] 38%|███▊ | 4155/10833 [00:38<00:55, 120.00it/s] 38%|███▊ | 4168/10833 [00:38<00:55, 119.96it/s] 39%|███▊ | 4180/10833 [00:38<00:55, 119.85it/s] 39%|███▊ | 4193/10833 [00:39<00:55, 119.93it/s] 39%|███▉ | 4205/10833 [00:39<00:55, 119.16it/s] 39%|███▉ | 4217/10833 [00:39<00:55, 119.31it/s] 39%|███▉ | 4229/10833 [00:39<00:55, 119.42it/s] 39%|███▉ | 4241/10833 [00:39<00:55, 119.53it/s] 39%|███▉ | 4253/10833 [00:39<00:55, 119.60it/s] 39%|███▉ | 4265/10833 [00:39<00:54, 119.70it/s] 39%|███▉ | 4277/10833 [00:39<00:54, 119.73it/s] 40%|███▉ | 4289/10833 [00:39<00:54, 119.73it/s] 40%|███▉ | 4301/10833 [00:39<00:54, 119.70it/s] 40%|███▉ | 4314/10833 [00:40<00:54, 119.88it/s] 40%|███▉ | 4327/10833 [00:40<00:54, 119.98it/s] 40%|████ | 4340/10833 [00:40<00:54, 119.98it/s] 40%|████ | 4353/10833 [00:40<00:54, 119.95it/s] 40%|████ | 4365/10833 [00:40<00:54, 119.10it/s] 40%|████ | 4377/10833 [00:40<00:54, 119.29it/s] 41%|████ | 4390/10833 [00:40<00:53, 119.53it/s] 41%|████ | 4402/10833 [00:40<00:53, 119.59it/s] 41%|████ | 4414/10833 [00:40<00:53, 119.56it/s] 41%|████ | 4427/10833 [00:41<00:53, 119.71it/s] 41%|████ | 4440/10833 [00:41<00:53, 119.82it/s] 41%|████ | 4453/10833 [00:41<00:53, 119.86it/s] 41%|████ | 4465/10833 [00:41<00:53, 119.76it/s] 41%|████▏ | 4477/10833 [00:41<00:53, 119.73it/s] 41%|████▏ | 4490/10833 [00:41<00:52, 119.83it/s] 42%|████▏ | 4502/10833 [00:41<00:52, 119.81it/s] 42%|████▏ | 4515/10833 [00:41<00:52, 119.91it/s] 42%|████▏ | 4527/10833 [00:41<00:52, 119.13it/s] 42%|████▏ | 4539/10833 [00:41<00:52, 119.38it/s] 42%|████▏ | 4552/10833 [00:42<00:52, 119.60it/s] 42%|████▏ | 4565/10833 [00:42<00:52, 119.73it/s] 42%|████▏ | 4577/10833 [00:42<00:52, 119.79it/s] 42%|████▏ | 4590/10833 [00:42<00:52, 119.96it/s] 42%|████▏ | 4603/10833 [00:42<00:51, 120.08it/s] 43%|████▎ | 4616/10833 [00:42<00:51, 120.22it/s] 43%|████▎ | 4629/10833 [00:42<00:51, 120.20it/s] 43%|████▎ | 4642/10833 [00:42<00:51, 120.34it/s] 43%|████▎ | 4655/10833 [00:42<00:51, 120.44it/s] 43%|████▎ | 4668/10833 [00:43<00:51, 120.33it/s] 43%|████▎ | 4681/10833 [00:43<00:51, 120.17it/s] 43%|████▎ | 4694/10833 [00:43<00:51, 119.47it/s] 43%|████▎ | 4707/10833 [00:43<00:51, 119.71it/s] 44%|████▎ | 4720/10833 [00:43<00:51, 119.85it/s] 44%|████▎ | 4733/10833 [00:43<00:50, 119.97it/s] 44%|████▍ | 4746/10833 [00:43<00:50, 120.01it/s] 44%|████▍ | 4759/10833 [00:43<00:50, 120.05it/s] 44%|████▍ | 4772/10833 [00:43<00:50, 120.08it/s] 44%|████▍ | 4785/10833 [00:44<00:50, 120.15it/s] 44%|████▍ | 4798/10833 [00:44<00:50, 120.23it/s] 44%|████▍ | 4811/10833 [00:44<00:50, 120.16it/s] 45%|████▍ | 4824/10833 [00:44<00:50, 120.16it/s] 45%|████▍ | 4837/10833 [00:44<00:49, 120.06it/s] 45%|████▍ | 4850/10833 [00:44<00:50, 119.30it/s] 45%|████▍ | 4863/10833 [00:44<00:49, 119.54it/s] 45%|████▌ | 4876/10833 [00:44<00:49, 119.77it/s] 45%|████▌ | 4889/10833 [00:44<00:49, 119.91it/s] 45%|████▌ | 4901/10833 [00:45<00:49, 119.93it/s] 45%|████▌ | 4914/10833 [00:45<00:49, 120.10it/s] 45%|████▌ | 4927/10833 [00:45<00:49, 120.16it/s] 46%|████▌ | 4940/10833 [00:45<00:49, 120.20it/s] 46%|████▌ | 4953/10833 [00:45<00:48, 120.10it/s] 46%|████▌ | 4966/10833 [00:45<00:48, 120.09it/s] 46%|████▌ | 4979/10833 [00:45<00:48, 120.01it/s] 46%|████▌ | 4992/10833 [00:45<00:48, 120.00it/s] 46%|████▌ | 5005/10833 [00:45<00:48, 119.34it/s] 46%|████▋ | 5018/10833 [00:45<00:48, 119.68it/s] 46%|████▋ | 5031/10833 [00:46<00:48, 119.80it/s] 47%|████▋ | 5044/10833 [00:46<00:48, 119.88it/s] 47%|████▋ | 5056/10833 [00:46<00:48, 119.88it/s] 47%|████▋ | 5069/10833 [00:46<00:48, 119.99it/s] 47%|████▋ | 5082/10833 [00:46<00:47, 120.15it/s] 47%|████▋ | 5095/10833 [00:46<00:47, 120.17it/s] 47%|████▋ | 5108/10833 [00:46<00:47, 120.13it/s] 47%|████▋ | 5121/10833 [00:46<00:47, 120.15it/s] 47%|████▋ | 5134/10833 [00:46<00:47, 120.18it/s] 48%|████▊ | 5147/10833 [00:47<00:47, 120.25it/s] 48%|████▊ | 5160/10833 [00:47<00:47, 120.14it/s] 48%|████▊ | 5173/10833 [00:47<00:47, 119.47it/s] 48%|████▊ | 5186/10833 [00:47<00:48, 117.02it/s] 48%|████▊ | 5199/10833 [00:47<00:47, 117.95it/s] 48%|████▊ | 5211/10833 [00:47<00:47, 118.41it/s] 48%|████▊ | 5224/10833 [00:47<00:47, 118.95it/s] 48%|████▊ | 5237/10833 [00:47<00:46, 119.28it/s] 48%|████▊ | 5249/10833 [00:47<00:46, 119.48it/s] 49%|████▊ | 5262/10833 [00:48<00:46, 119.69it/s] 49%|████▊ | 5275/10833 [00:48<00:46, 119.82it/s] 49%|████▉ | 5288/10833 [00:48<00:46, 119.92it/s] 49%|████▉ | 5300/10833 [00:48<00:46, 119.93it/s] 49%|████▉ | 5313/10833 [00:48<00:45, 120.04it/s] 49%|████▉ | 5326/10833 [00:48<00:46, 119.30it/s] 49%|████▉ | 5339/10833 [00:48<00:45, 119.53it/s] 49%|████▉ | 5351/10833 [00:48<00:45, 119.62it/s] 50%|████▉ | 5363/10833 [00:48<00:45, 119.72it/s] 50%|████▉ | 5375/10833 [00:48<00:45, 119.76it/s] 50%|████▉ | 5388/10833 [00:49<00:45, 119.88it/s] 50%|████▉ | 5401/10833 [00:49<00:45, 119.90it/s] 50%|████▉ | 5413/10833 [00:49<00:45, 119.90it/s] 50%|█████ | 5426/10833 [00:49<00:45, 119.95it/s] 50%|█████ | 5439/10833 [00:49<00:44, 119.96it/s] 50%|█████ | 5452/10833 [00:49<00:44, 120.04it/s] 50%|█████ | 5465/10833 [00:49<00:44, 120.08it/s] 51%|█████ | 5478/10833 [00:49<00:44, 120.05it/s] 51%|█████ | 5491/10833 [00:49<00:44, 119.36it/s] 51%|█████ | 5504/10833 [00:50<00:44, 119.57it/s] 51%|█████ | 5517/10833 [00:50<00:44, 119.91it/s] 51%|█████ | 5530/10833 [00:50<00:44, 120.06it/s] 51%|█████ | 5543/10833 [00:50<00:44, 120.10it/s] 51%|█████▏ | 5556/10833 [00:50<00:43, 120.12it/s] 51%|█████▏ | 5569/10833 [00:50<00:43, 120.18it/s] 52%|█████▏ | 5582/10833 [00:50<00:43, 120.18it/s] 52%|█████▏ | 5595/10833 [00:50<00:43, 120.24it/s] 52%|█████▏ | 5608/10833 [00:50<00:43, 120.33it/s] 52%|█████▏ | 5621/10833 [00:51<00:43, 120.26it/s] 52%|█████▏ | 5634/10833 [00:51<00:43, 120.30it/s] 52%|█████▏ | 5647/10833 [00:51<00:43, 119.51it/s] 52%|█████▏ | 5660/10833 [00:51<00:43, 119.65it/s] 52%|█████▏ | 5673/10833 [00:51<00:43, 119.79it/s] 52%|█████▏ | 5686/10833 [00:51<00:42, 119.95it/s] 53%|█████▎ | 5698/10833 [00:51<00:42, 119.88it/s] 53%|█████▎ | 5710/10833 [00:51<00:42, 119.88it/s] 53%|█████▎ | 5723/10833 [00:51<00:42, 119.89it/s] 53%|█████▎ | 5736/10833 [00:51<00:42, 120.00it/s] 53%|█████▎ | 5748/10833 [00:52<00:42, 119.91it/s] 53%|█████▎ | 5761/10833 [00:52<00:42, 120.03it/s] 53%|█████▎ | 5774/10833 [00:52<00:42, 120.06it/s] 53%|█████▎ | 5787/10833 [00:52<00:42, 120.09it/s] 54%|█████▎ | 5800/10833 [00:52<00:41, 120.15it/s] 54%|█████▎ | 5813/10833 [00:52<00:41, 119.55it/s] 54%|█████▍ | 5826/10833 [00:52<00:41, 119.71it/s] 54%|█████▍ | 5838/10833 [00:52<00:41, 119.73it/s] 54%|█████▍ | 5851/10833 [00:52<00:41, 119.84it/s] 54%|█████▍ | 5864/10833 [00:53<00:41, 119.91it/s] 54%|█████▍ | 5876/10833 [00:53<00:41, 119.91it/s] 54%|█████▍ | 5889/10833 [00:53<00:41, 119.98it/s] 54%|█████▍ | 5902/10833 [00:53<00:41, 120.02it/s] 55%|█████▍ | 5915/10833 [00:53<00:40, 120.04it/s] 55%|█████▍ | 5928/10833 [00:53<00:40, 120.16it/s] 55%|█████▍ | 5941/10833 [00:53<00:40, 120.18it/s] 55%|█████▍ | 5954/10833 [00:53<00:40, 120.13it/s] 55%|█████▌ | 5967/10833 [00:53<00:40, 120.12it/s] 55%|█████▌ | 5980/10833 [00:54<00:40, 119.53it/s] 55%|█████▌ | 5992/10833 [00:54<00:40, 119.64it/s] 55%|█████▌ | 6005/10833 [00:54<00:40, 119.84it/s] 56%|█████▌ | 6017/10833 [00:54<00:40, 119.88it/s] 56%|█████▌ | 6030/10833 [00:54<00:40, 119.97it/s] 56%|█████▌ | 6043/10833 [00:54<00:39, 120.13it/s] 56%|█████▌ | 6056/10833 [00:54<00:39, 120.12it/s] 56%|█████▌ | 6069/10833 [00:54<00:39, 120.13it/s] 56%|█████▌ | 6082/10833 [00:54<00:39, 120.12it/s] 56%|█████▋ | 6095/10833 [00:54<00:39, 120.15it/s] 56%|█████▋ | 6108/10833 [00:55<00:39, 120.15it/s] 57%|█████▋ | 6121/10833 [00:55<00:39, 120.04it/s] 57%|█████▋ | 6134/10833 [00:55<00:39, 119.39it/s] 57%|█████▋ | 6147/10833 [00:55<00:39, 119.61it/s] 57%|█████▋ | 6160/10833 [00:55<00:38, 119.82it/s] 57%|█████▋ | 6173/10833 [00:55<00:38, 119.89it/s] 57%|█████▋ | 6186/10833 [00:55<00:38, 119.98it/s] 57%|█████▋ | 6198/10833 [00:55<00:38, 119.98it/s] 57%|█████▋ | 6210/10833 [00:55<00:38, 119.95it/s] 57%|█████▋ | 6222/10833 [00:56<00:38, 119.96it/s] 58%|█████▊ | 6234/10833 [00:56<00:38, 119.95it/s] 58%|█████▊ | 6247/10833 [00:56<00:38, 120.11it/s] 58%|█████▊ | 6260/10833 [00:56<00:38, 120.13it/s] 58%|█████▊ | 6273/10833 [00:56<00:37, 120.19it/s] 58%|█████▊ | 6286/10833 [00:56<00:37, 120.10it/s] 58%|█████▊ | 6299/10833 [00:56<00:37, 119.46it/s] 58%|█████▊ | 6312/10833 [00:56<00:37, 119.67it/s] 58%|█████▊ | 6324/10833 [00:56<00:37, 119.75it/s] 58%|█████▊ | 6337/10833 [00:56<00:37, 119.90it/s] 59%|█████▊ | 6349/10833 [00:57<00:37, 119.91it/s] 59%|█████▊ | 6361/10833 [00:57<00:37, 119.92it/s] 59%|█████▉ | 6374/10833 [00:57<00:37, 119.98it/s] 59%|█████▉ | 6386/10833 [00:57<00:37, 119.86it/s] 59%|█████▉ | 6399/10833 [00:57<00:36, 120.01it/s] 59%|█████▉ | 6412/10833 [00:57<00:36, 120.06it/s] 59%|█████▉ | 6425/10833 [00:57<00:36, 120.04it/s] 59%|█████▉ | 6438/10833 [00:57<00:36, 120.02it/s] 60%|█████▉ | 6451/10833 [00:57<00:36, 120.06it/s] 60%|█████▉ | 6464/10833 [00:58<00:36, 119.28it/s] 60%|█████▉ | 6477/10833 [00:58<00:36, 119.55it/s] 60%|█████▉ | 6489/10833 [00:58<00:36, 119.61it/s] 60%|██████ | 6502/10833 [00:58<00:36, 119.81it/s] 60%|██████ | 6514/10833 [00:58<00:36, 119.85it/s] 60%|██████ | 6527/10833 [00:58<00:35, 119.90it/s] 60%|██████ | 6540/10833 [00:58<00:35, 120.01it/s] 60%|██████ | 6553/10833 [00:58<00:35, 120.08it/s] 61%|██████ | 6566/10833 [00:58<00:35, 120.13it/s] 61%|██████ | 6579/10833 [00:59<00:35, 120.11it/s] 61%|██████ | 6592/10833 [00:59<00:35, 120.13it/s] 61%|██████ | 6605/10833 [00:59<00:35, 120.02it/s] 61%|██████ | 6618/10833 [00:59<00:35, 119.36it/s] 61%|██████ | 6631/10833 [00:59<00:35, 119.65it/s] 61%|██████▏ | 6644/10833 [00:59<00:34, 119.88it/s] 61%|██████▏ | 6657/10833 [00:59<00:34, 120.02it/s] 62%|██████▏ | 6670/10833 [00:59<00:34, 120.11it/s] 62%|██████▏ | 6683/10833 [00:59<00:34, 120.16it/s] 62%|██████▏ | 6696/10833 [00:59<00:34, 120.17it/s] 62%|██████▏ | 6709/10833 [01:00<00:34, 120.21it/s] 62%|██████▏ | 6722/10833 [01:00<00:34, 120.30it/s] 62%|██████▏ | 6735/10833 [01:00<00:34, 120.26it/s] 62%|██████▏ | 6748/10833 [01:00<00:33, 120.28it/s] 62%|██████▏ | 6761/10833 [01:00<00:33, 120.26it/s] 63%|██████▎ | 6774/10833 [01:00<00:33, 119.58it/s] 63%|██████▎ | 6787/10833 [01:00<00:33, 119.77it/s] 63%|██████▎ | 6800/10833 [01:00<00:33, 119.95it/s] 63%|██████▎ | 6812/10833 [01:00<00:33, 119.95it/s] 63%|██████▎ | 6824/10833 [01:01<00:33, 119.92it/s] 63%|██████▎ | 6837/10833 [01:01<00:33, 120.00it/s] 63%|██████▎ | 6850/10833 [01:01<00:33, 120.07it/s] 63%|██████▎ | 6863/10833 [01:01<00:33, 120.14it/s] 63%|██████▎ | 6876/10833 [01:01<00:32, 120.11it/s] 64%|██████▎ | 6889/10833 [01:01<00:32, 120.16it/s] 64%|██████▎ | 6902/10833 [01:01<00:32, 120.17it/s] 64%|██████▍ | 6915/10833 [01:01<00:32, 120.22it/s] 64%|██████▍ | 6928/10833 [01:01<00:32, 120.14it/s] 64%|██████▍ | 6941/10833 [01:02<00:32, 119.40it/s] 64%|██████▍ | 6954/10833 [01:02<00:32, 119.65it/s] 64%|██████▍ | 6967/10833 [01:02<00:32, 119.82it/s] 64%|██████▍ | 6980/10833 [01:02<00:32, 120.00it/s] 65%|██████▍ | 6993/10833 [01:02<00:32, 119.98it/s] 65%|██████▍ | 7006/10833 [01:02<00:31, 120.05it/s] 65%|██████▍ | 7019/10833 [01:02<00:31, 120.00it/s] 65%|██████▍ | 7032/10833 [01:02<00:31, 120.08it/s] 65%|██████▌ | 7045/10833 [01:02<00:31, 120.16it/s] 65%|██████▌ | 7058/10833 [01:02<00:31, 120.03it/s] 65%|██████▌ | 7071/10833 [01:03<00:31, 120.02it/s] 65%|██████▌ | 7084/10833 [01:03<00:31, 120.06it/s] 66%|██████▌ | 7097/10833 [01:03<00:31, 119.32it/s] 66%|██████▌ | 7109/10833 [01:03<00:31, 119.47it/s] 66%|██████▌ | 7121/10833 [01:03<00:31, 119.55it/s] 66%|██████▌ | 7134/10833 [01:03<00:30, 119.79it/s] 66%|██████▌ | 7146/10833 [01:03<00:30, 119.85it/s] 66%|██████▌ | 7159/10833 [01:03<00:30, 120.02it/s] 66%|██████▌ | 7172/10833 [01:03<00:30, 120.15it/s] 66%|██████▋ | 7185/10833 [01:04<00:30, 120.18it/s] 66%|██████▋ | 7198/10833 [01:04<00:30, 120.20it/s] 67%|██████▋ | 7211/10833 [01:04<00:30, 120.18it/s] 67%|██████▋ | 7224/10833 [01:04<00:30, 120.22it/s] 67%|██████▋ | 7237/10833 [01:04<00:29, 120.20it/s] 67%|██████▋ | 7250/10833 [01:04<00:29, 120.27it/s] 67%|██████▋ | 7263/10833 [01:04<00:30, 117.01it/s] 67%|██████▋ | 7276/10833 [01:04<00:30, 118.03it/s] 67%|██████▋ | 7289/10833 [01:04<00:29, 118.71it/s] 67%|██████▋ | 7302/10833 [01:05<00:29, 119.21it/s] 68%|██████▊ | 7314/10833 [01:05<00:29, 119.39it/s] 68%|██████▊ | 7327/10833 [01:05<00:29, 119.59it/s] 68%|██████▊ | 7340/10833 [01:05<00:29, 119.80it/s] 68%|██████▊ | 7353/10833 [01:05<00:29, 119.90it/s] 68%|██████▊ | 7365/10833 [01:05<00:28, 119.88it/s] 68%|██████▊ | 7378/10833 [01:05<00:28, 120.02it/s] 68%|██████▊ | 7391/10833 [01:05<00:28, 120.07it/s] 68%|██████▊ | 7404/10833 [01:05<00:28, 120.10it/s] 68%|██████▊ | 7417/10833 [01:05<00:28, 119.38it/s] 69%|██████▊ | 7430/10833 [01:06<00:28, 119.69it/s] 69%|██████▊ | 7443/10833 [01:06<00:28, 119.82it/s] 69%|██████▉ | 7456/10833 [01:06<00:28, 119.95it/s] 69%|██████▉ | 7469/10833 [01:06<00:28, 119.99it/s] 69%|██████▉ | 7482/10833 [01:06<00:27, 120.08it/s] 69%|██████▉ | 7495/10833 [01:06<00:27, 120.11it/s] 69%|██████▉ | 7508/10833 [01:06<00:27, 120.12it/s] 69%|██████▉ | 7521/10833 [01:06<00:27, 120.08it/s] 70%|██████▉ | 7534/10833 [01:06<00:27, 120.03it/s] 70%|██████▉ | 7547/10833 [01:07<00:27, 120.10it/s] 70%|██████▉ | 7560/10833 [01:07<00:27, 120.07it/s] 70%|██████▉ | 7573/10833 [01:07<00:27, 120.10it/s] 70%|███████ | 7586/10833 [01:07<00:27, 119.36it/s] 70%|███████ | 7599/10833 [01:07<00:27, 119.56it/s] 70%|███████ | 7612/10833 [01:07<00:26, 119.73it/s] 70%|███████ | 7625/10833 [01:07<00:26, 119.87it/s] 71%|███████ | 7638/10833 [01:07<00:26, 119.99it/s] 71%|███████ | 7650/10833 [01:07<00:26, 119.94it/s] 71%|███████ | 7662/10833 [01:08<00:26, 119.95it/s] 71%|███████ | 7674/10833 [01:08<00:26, 119.95it/s] 71%|███████ | 7686/10833 [01:08<00:26, 119.93it/s] 71%|███████ | 7699/10833 [01:08<00:26, 120.10it/s] 71%|███████ | 7712/10833 [01:08<00:25, 120.12it/s] 71%|███████▏ | 7725/10833 [01:08<00:25, 120.20it/s] 71%|███████▏ | 7738/10833 [01:08<00:25, 119.36it/s] 72%|███████▏ | 7751/10833 [01:08<00:25, 119.67it/s] 72%|███████▏ | 7763/10833 [01:08<00:25, 119.75it/s] 72%|███████▏ | 7775/10833 [01:08<00:25, 119.82it/s] 72%|███████▏ | 7788/10833 [01:09<00:25, 119.89it/s] 72%|███████▏ | 7801/10833 [01:09<00:25, 120.07it/s] 72%|███████▏ | 7814/10833 [01:09<00:25, 120.04it/s] 72%|███████▏ | 7827/10833 [01:09<00:25, 120.11it/s] 72%|███████▏ | 7840/10833 [01:09<00:24, 120.09it/s] 72%|███████▏ | 7853/10833 [01:09<00:24, 120.06it/s] 73%|███████▎ | 7866/10833 [01:09<00:24, 120.07it/s] 73%|███████▎ | 7879/10833 [01:09<00:24, 120.19it/s] 73%|███████▎ | 7892/10833 [01:09<00:24, 120.15it/s] 73%|███████▎ | 7905/10833 [01:10<00:24, 119.37it/s] 73%|███████▎ | 7918/10833 [01:10<00:24, 119.71it/s] 73%|███████▎ | 7931/10833 [01:10<00:24, 119.90it/s] 73%|███████▎ | 7943/10833 [01:10<00:24, 119.92it/s] 73%|███████▎ | 7956/10833 [01:10<00:23, 119.99it/s] 74%|███████▎ | 7968/10833 [01:10<00:23, 119.95it/s] 74%|███████▎ | 7981/10833 [01:10<00:23, 120.03it/s] 74%|███████▍ | 7994/10833 [01:10<00:23, 120.15it/s] 74%|███████▍ | 8007/10833 [01:10<00:23, 120.15it/s] 74%|███████▍ | 8020/10833 [01:11<00:23, 119.92it/s] 74%|███████▍ | 8033/10833 [01:11<00:23, 119.97it/s] 74%|███████▍ | 8045/10833 [01:11<00:23, 119.93it/s] 74%|███████▍ | 8058/10833 [01:11<00:23, 119.99it/s] 74%|███████▍ | 8070/10833 [01:11<00:23, 119.40it/s] 75%|███████▍ | 8082/10833 [01:11<00:23, 119.47it/s] 75%|███████▍ | 8094/10833 [01:11<00:22, 119.60it/s] 75%|███████▍ | 8107/10833 [01:11<00:22, 119.75it/s] 75%|███████▍ | 8120/10833 [01:11<00:22, 119.93it/s] 75%|███████▌ | 8133/10833 [01:11<00:22, 119.96it/s] 75%|███████▌ | 8146/10833 [01:12<00:22, 120.02it/s] 75%|███████▌ | 8159/10833 [01:12<00:22, 120.00it/s] 75%|███████▌ | 8172/10833 [01:12<00:22, 120.08it/s] 76%|███████▌ | 8185/10833 [01:12<00:22, 120.15it/s] 76%|███████▌ | 8198/10833 [01:12<00:21, 120.17it/s] 76%|███████▌ | 8211/10833 [01:12<00:21, 120.04it/s] 76%|███████▌ | 8224/10833 [01:12<00:21, 119.44it/s] 76%|███████▌ | 8237/10833 [01:12<00:21, 119.65it/s] 76%|███████▌ | 8249/10833 [01:12<00:21, 119.70it/s] 76%|███████▋ | 8262/10833 [01:13<00:21, 119.85it/s] 76%|███████▋ | 8275/10833 [01:13<00:21, 119.96it/s] 77%|███████▋ | 8288/10833 [01:13<00:21, 120.03it/s] 77%|███████▋ | 8301/10833 [01:13<00:21, 119.94it/s] 77%|███████▋ | 8314/10833 [01:13<00:20, 120.06it/s] 77%|███████▋ | 8327/10833 [01:13<00:20, 120.13it/s] 77%|███████▋ | 8340/10833 [01:13<00:20, 120.14it/s] 77%|███████▋ | 8353/10833 [01:13<00:20, 120.07it/s] 77%|███████▋ | 8366/10833 [01:13<00:20, 120.18it/s] 77%|███████▋ | 8379/10833 [01:14<00:20, 120.26it/s] 77%|███████▋ | 8392/10833 [01:14<00:20, 119.45it/s] 78%|███████▊ | 8404/10833 [01:14<00:20, 119.51it/s] 78%|███████▊ | 8416/10833 [01:14<00:20, 119.60it/s] 78%|███████▊ | 8429/10833 [01:14<00:20, 119.79it/s] 78%|███████▊ | 8442/10833 [01:14<00:19, 119.90it/s] 78%|███████▊ | 8455/10833 [01:14<00:19, 120.00it/s] 78%|███████▊ | 8467/10833 [01:14<00:19, 119.97it/s] 78%|███████▊ | 8479/10833 [01:14<00:19, 119.90it/s] 78%|███████▊ | 8491/10833 [01:14<00:19, 119.92it/s] 78%|███████▊ | 8503/10833 [01:15<00:19, 119.93it/s] 79%|███████▊ | 8515/10833 [01:15<00:19, 119.92it/s] 79%|███████▊ | 8528/10833 [01:15<00:19, 120.01it/s] 79%|███████▉ | 8541/10833 [01:15<00:19, 119.92it/s] 79%|███████▉ | 8553/10833 [01:15<00:19, 119.23it/s] 79%|███████▉ | 8566/10833 [01:15<00:18, 119.52it/s] 79%|███████▉ | 8578/10833 [01:15<00:18, 119.62it/s] 79%|███████▉ | 8591/10833 [01:15<00:18, 119.75it/s] 79%|███████▉ | 8604/10833 [01:15<00:18, 119.88it/s] 80%|███████▉ | 8616/10833 [01:15<00:18, 119.89it/s] 80%|███████▉ | 8629/10833 [01:16<00:18, 120.00it/s] 80%|███████▉ | 8642/10833 [01:16<00:18, 120.02it/s] 80%|███████▉ | 8655/10833 [01:16<00:18, 120.10it/s] 80%|████████ | 8668/10833 [01:16<00:18, 120.16it/s] 80%|████████ | 8681/10833 [01:16<00:17, 120.17it/s] 80%|████████ | 8694/10833 [01:16<00:17, 120.13it/s] 80%|████████ | 8707/10833 [01:16<00:17, 119.45it/s] 80%|████████ | 8720/10833 [01:16<00:17, 119.73it/s] 81%|████████ | 8733/10833 [01:16<00:17, 119.87it/s] 81%|████████ | 8746/10833 [01:17<00:17, 119.98it/s] 81%|████████ | 8759/10833 [01:17<00:17, 120.06it/s] 81%|████████ | 8772/10833 [01:17<00:17, 120.06it/s] 81%|████████ | 8785/10833 [01:17<00:17, 120.14it/s] 81%|████████ | 8798/10833 [01:17<00:16, 120.17it/s] 81%|████████▏ | 8811/10833 [01:17<00:16, 120.19it/s] 81%|████████▏ | 8824/10833 [01:17<00:16, 120.11it/s] 82%|████████▏ | 8837/10833 [01:17<00:16, 120.07it/s] 82%|████████▏ | 8850/10833 [01:17<00:16, 120.13it/s] 82%|████████▏ | 8863/10833 [01:18<00:16, 120.22it/s] 82%|████████▏ | 8876/10833 [01:18<00:16, 119.51it/s] 82%|████████▏ | 8889/10833 [01:18<00:16, 119.68it/s] 82%|████████▏ | 8902/10833 [01:18<00:16, 119.82it/s] 82%|████████▏ | 8915/10833 [01:18<00:15, 119.89it/s] 82%|████████▏ | 8928/10833 [01:18<00:15, 119.97it/s] 83%|████████▎ | 8940/10833 [01:18<00:15, 119.96it/s] 83%|████████▎ | 8953/10833 [01:18<00:15, 120.04it/s] 83%|████████▎ | 8966/10833 [01:18<00:15, 120.09it/s] 83%|████████▎ | 8979/10833 [01:19<00:15, 120.09it/s] 83%|████████▎ | 8992/10833 [01:19<00:15, 120.00it/s] 83%|████████▎ | 9004/10833 [01:19<00:15, 119.95it/s] 83%|████████▎ | 9017/10833 [01:19<00:15, 119.97it/s] 83%|████████▎ | 9029/10833 [01:19<00:15, 119.21it/s] 83%|████████▎ | 9042/10833 [01:19<00:14, 119.53it/s] 84%|████████▎ | 9055/10833 [01:19<00:14, 119.77it/s] 84%|████████▎ | 9068/10833 [01:19<00:14, 119.91it/s] 84%|████████▍ | 9081/10833 [01:19<00:14, 119.96it/s] 84%|████████▍ | 9094/10833 [01:19<00:14, 119.98it/s] 84%|████████▍ | 9107/10833 [01:20<00:14, 120.01it/s] 84%|████████▍ | 9120/10833 [01:20<00:14, 120.14it/s] 84%|████████▍ | 9133/10833 [01:20<00:14, 120.21it/s] 84%|████████▍ | 9146/10833 [01:20<00:14, 120.22it/s] 85%|████████▍ | 9159/10833 [01:20<00:13, 120.18it/s] 85%|████████▍ | 9172/10833 [01:20<00:13, 120.17it/s] 85%|████████▍ | 9185/10833 [01:20<00:13, 119.47it/s] 85%|████████▍ | 9198/10833 [01:20<00:13, 119.67it/s] 85%|████████▌ | 9211/10833 [01:20<00:13, 119.91it/s] 85%|████████▌ | 9224/10833 [01:21<00:13, 119.95it/s] 85%|████████▌ | 9237/10833 [01:21<00:13, 120.09it/s] 85%|████████▌ | 9250/10833 [01:21<00:13, 120.05it/s] 86%|████████▌ | 9263/10833 [01:21<00:13, 120.08it/s] 86%|████████▌ | 9276/10833 [01:21<00:12, 120.07it/s] 86%|████████▌ | 9289/10833 [01:21<00:12, 120.13it/s] 86%|████████▌ | 9302/10833 [01:21<00:12, 120.07it/s] 86%|████████▌ | 9315/10833 [01:21<00:12, 120.10it/s] 86%|████████▌ | 9328/10833 [01:21<00:12, 120.08it/s] 86%|████████▌ | 9341/10833 [01:22<00:12, 117.47it/s] 86%|████████▋ | 9353/10833 [01:22<00:12, 117.50it/s] 86%|████████▋ | 9365/10833 [01:22<00:12, 118.16it/s] 87%|████████▋ | 9378/10833 [01:22<00:12, 118.85it/s] 87%|████████▋ | 9391/10833 [01:22<00:12, 119.26it/s] 87%|████████▋ | 9404/10833 [01:22<00:11, 119.56it/s] 87%|████████▋ | 9417/10833 [01:22<00:11, 119.74it/s] 87%|████████▋ | 9430/10833 [01:22<00:11, 119.86it/s] 87%|████████▋ | 9443/10833 [01:22<00:11, 119.99it/s] 87%|████████▋ | 9456/10833 [01:23<00:11, 120.08it/s] 87%|████████▋ | 9469/10833 [01:23<00:11, 120.13it/s] 88%|████████▊ | 9482/10833 [01:23<00:11, 120.15it/s] 88%|████████▊ | 9495/10833 [01:23<00:11, 120.15it/s] 88%|████████▊ | 9508/10833 [01:23<00:11, 119.42it/s] 88%|████████▊ | 9521/10833 [01:23<00:10, 119.67it/s] 88%|████████▊ | 9534/10833 [01:23<00:10, 119.82it/s] 88%|████████▊ | 9546/10833 [01:23<00:10, 119.83it/s] 88%|████████▊ | 9558/10833 [01:23<00:10, 119.84it/s] 88%|████████▊ | 9570/10833 [01:23<00:10, 119.86it/s] 88%|████████▊ | 9583/10833 [01:24<00:10, 119.95it/s] 89%|████████▊ | 9596/10833 [01:24<00:10, 120.00it/s] 89%|████████▊ | 9609/10833 [01:24<00:10, 120.02it/s] 89%|████████▉ | 9622/10833 [01:24<00:10, 120.17it/s] 89%|████████▉ | 9635/10833 [01:24<00:09, 120.23it/s] 89%|████████▉ | 9648/10833 [01:24<00:09, 120.22it/s] 89%|████████▉ | 9661/10833 [01:24<00:09, 120.26it/s] 89%|████████▉ | 9674/10833 [01:24<00:09, 119.59it/s] 89%|████████▉ | 9687/10833 [01:24<00:09, 119.87it/s] 90%|████████▉ | 9699/10833 [01:25<00:09, 119.85it/s] 90%|████████▉ | 9712/10833 [01:25<00:09, 119.95it/s] 90%|████████▉ | 9725/10833 [01:25<00:09, 120.04it/s] 90%|████████▉ | 9738/10833 [01:25<00:09, 120.10it/s] 90%|█████████ | 9751/10833 [01:25<00:09, 120.11it/s] 90%|█████████ | 9764/10833 [01:25<00:08, 120.01it/s] 90%|█████████ | 9777/10833 [01:25<00:08, 120.07it/s] 90%|█████████ | 9790/10833 [01:25<00:08, 120.07it/s] 90%|█████████ | 9803/10833 [01:25<00:08, 120.08it/s] 91%|█████████ | 9816/10833 [01:26<00:08, 120.18it/s] 91%|█████████ | 9829/10833 [01:26<00:08, 119.39it/s] 91%|█████████ | 9842/10833 [01:26<00:08, 119.65it/s] 91%|█████████ | 9855/10833 [01:26<00:08, 119.85it/s] 91%|█████████ | 9867/10833 [01:26<00:08, 119.88it/s] 91%|█████████ | 9880/10833 [01:26<00:07, 119.97it/s] 91%|█████████▏| 9892/10833 [01:26<00:07, 119.93it/s] 91%|█████████▏| 9905/10833 [01:26<00:07, 120.06it/s] 92%|█████████▏| 9918/10833 [01:26<00:07, 119.98it/s] 92%|█████████▏| 9931/10833 [01:26<00:07, 119.95it/s] 92%|█████████▏| 9944/10833 [01:27<00:07, 120.08it/s] 92%|█████████▏| 9957/10833 [01:27<00:07, 120.12it/s] 92%|█████████▏| 9970/10833 [01:27<00:07, 120.14it/s] 92%|█████████▏| 9983/10833 [01:27<00:07, 120.14it/s] 92%|█████████▏| 9996/10833 [01:27<00:07, 119.40it/s] 92%|█████████▏| 10008/10833 [01:27<00:06, 119.54it/s] 93%|█████████▎| 10021/10833 [01:27<00:06, 119.77it/s] 93%|█████████▎| 10034/10833 [01:27<00:06, 119.86it/s] 93%|█████████▎| 10047/10833 [01:27<00:06, 119.93it/s] 93%|█████████▎| 10059/10833 [01:28<00:06, 119.90it/s] 93%|█████████▎| 10071/10833 [01:28<00:06, 119.87it/s] 93%|█████████▎| 10083/10833 [01:28<00:06, 119.91it/s] 93%|█████████▎| 10095/10833 [01:28<00:06, 119.82it/s] 93%|█████████▎| 10108/10833 [01:28<00:06, 119.93it/s] 93%|█████████▎| 10121/10833 [01:28<00:05, 120.07it/s] 94%|█████████▎| 10134/10833 [01:28<00:05, 120.05it/s] 94%|█████████▎| 10147/10833 [01:28<00:05, 120.06it/s] 94%|█████████▍| 10160/10833 [01:28<00:05, 119.34it/s] 94%|█████████▍| 10172/10833 [01:28<00:05, 119.50it/s] 94%|█████████▍| 10185/10833 [01:29<00:05, 119.71it/s] 94%|█████████▍| 10198/10833 [01:29<00:05, 119.88it/s] 94%|█████████▍| 10210/10833 [01:29<00:05, 119.87it/s] 94%|█████████▍| 10223/10833 [01:29<00:05, 120.06it/s] 94%|█████████▍| 10236/10833 [01:29<00:04, 120.03it/s] 95%|█████████▍| 10249/10833 [01:29<00:04, 119.92it/s] 95%|█████████▍| 10261/10833 [01:29<00:04, 119.91it/s] 95%|█████████▍| 10274/10833 [01:29<00:04, 120.08it/s] 95%|█████████▍| 10287/10833 [01:29<00:04, 120.19it/s] 95%|█████████▌| 10300/10833 [01:30<00:04, 120.18it/s] 95%|█████████▌| 10313/10833 [01:30<00:04, 119.38it/s] 95%|█████████▌| 10325/10833 [01:30<00:04, 119.53it/s] 95%|█████████▌| 10338/10833 [01:30<00:04, 119.77it/s] 96%|█████████▌| 10351/10833 [01:30<00:04, 119.86it/s] 96%|█████████▌| 10364/10833 [01:30<00:03, 120.00it/s] 96%|█████████▌| 10376/10833 [01:30<00:03, 119.83it/s] 96%|█████████▌| 10388/10833 [01:30<00:03, 119.80it/s] 96%|█████████▌| 10401/10833 [01:30<00:03, 119.98it/s] 96%|█████████▌| 10413/10833 [01:30<00:03, 119.94it/s] 96%|█████████▌| 10425/10833 [01:31<00:03, 119.92it/s] 96%|█████████▋| 10437/10833 [01:31<00:03, 119.92it/s] 96%|█████████▋| 10450/10833 [01:31<00:03, 120.05it/s] 97%|█████████▋| 10463/10833 [01:31<00:03, 120.08it/s] 97%|█████████▋| 10476/10833 [01:31<00:02, 119.33it/s] 97%|█████████▋| 10489/10833 [01:31<00:02, 119.59it/s] 97%|█████████▋| 10501/10833 [01:31<00:02, 119.69it/s] 97%|█████████▋| 10513/10833 [01:31<00:02, 119.74it/s] 97%|█████████▋| 10526/10833 [01:31<00:02, 119.84it/s] 97%|█████████▋| 10539/10833 [01:32<00:02, 119.96it/s] 97%|█████████▋| 10551/10833 [01:32<00:02, 119.96it/s] 98%|█████████▊| 10564/10833 [01:32<00:02, 119.96it/s] 98%|█████████▊| 10577/10833 [01:32<00:02, 120.05it/s] 98%|█████████▊| 10590/10833 [01:32<00:02, 120.06it/s] 98%|█████████▊| 10603/10833 [01:32<00:01, 120.10it/s] 98%|█████████▊| 10616/10833 [01:32<00:01, 120.02it/s] 98%|█████████▊| 10629/10833 [01:32<00:01, 119.96it/s] 98%|█████████▊| 10641/10833 [01:32<00:01, 119.17it/s] 98%|█████████▊| 10653/10833 [01:32<00:01, 119.35it/s] 98%|█████████▊| 10666/10833 [01:33<00:01, 119.55it/s] 99%|█████████▊| 10678/10833 [01:33<00:01, 119.63it/s] 99%|█████████▊| 10690/10833 [01:33<00:01, 119.64it/s] 99%|█████████▉| 10703/10833 [01:33<00:01, 119.77it/s] 99%|█████████▉| 10716/10833 [01:33<00:00, 119.86it/s] 99%|█████████▉| 10729/10833 [01:33<00:00, 119.91it/s] 99%|█████████▉| 10741/10833 [01:33<00:00, 119.90it/s] 99%|█████████▉| 10753/10833 [01:33<00:00, 119.91it/s] 99%|█████████▉| 10766/10833 [01:33<00:00, 120.04it/s] 100%|█████████▉| 10779/10833 [01:34<00:00, 120.13it/s] 100%|█████████▉| 10792/10833 [01:34<00:00, 119.42it/s] 100%|█████████▉| 10805/10833 [01:34<00:00, 119.66it/s] 100%|█████████▉| 10818/10833 [01:34<00:00, 119.86it/s] 100%|█████████▉| 10831/10833 [01:34<00:00, 119.99it/s] 100%|██████████| 10833/10833 [01:34<00:00, 114.64it/s]
INFO:TestAccuracy:Batch size is 1, F1: 85.85489, Exact Match:78.75012
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
INFO:BackendDCU:Batch size is 1, QPS: 126, Avg Latency:7.91, Tail Latency:10.12
INFO:BackendDCU:Batch size is 2, QPS: 147, Avg Latency:13.54, Tail Latency:15.56
INFO:BackendDCU:Batch size is 4, QPS: 179, Avg Latency:22.32, Tail Latency:24.36
INFO:BackendDCU:Batch size is 8, QPS: 190, Avg Latency:42.07, Tail Latency:43.95
INFO:BackendDCU:Batch size is 16, QPS: 197, Avg Latency:80.9, Tail Latency:83.42
INFO:BackendDCU:Batch size is 32, QPS: 220, Avg Latency:145.28, Tail Latency:147.23
INFO:BackendDCU:Batch size is 64, QPS: 223, Avg Latency:286.53, Tail Latency:288.91
INFO:BackendDCU:Batch size is 128, QPS: 209, Avg Latency:609.58, Tail Latency:612.16
INFO:BackendDCU:Batch size is 256, QPS: 208, Avg Latency:1228.8, Tail Latency:1231.68
INFO:BackendDCU:Batch size is 512, QPS: 207, Avg Latency:2470.29, Tail Latency:2474.04
INFO:BackendDCU:Batch size is 1024, QPS: 206, Avg Latency:4960.97, Tail Latency:4969.59
INFO:PerfEngine:Testing Finish. Report is saved in path: [ general_perf/reports/DCU/bert-torch-fp32/result-fp32.json ]
INFO:PerfEngine:PDF Version is saved in path: [ general_perf/reports/DCU/bert-torch-fp32/BERT-TORCH-FP32-TO-FP32.JSON.pdf ]
Writing predictions to: /home/workspace/ByteMLPerf/byte_infer_perf/general_perf/reports/DCU/predictions.json
ByteMLPerf/byte_infer_perf/general_perf/log/clip-onnx-fp32.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
INFO:PerfEngine:******************************************* Start to test model: clip-onnx-fp32. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: fake_dataset
INFO:PerfEngine:Start to compile the model...
2024-11-13 16:58:27.053598215 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:58:27.053618792 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
2024-11-13 16:58:33.081943533 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:58:33.081959626 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 4, QPS: 715, Avg Latency:5.59, Tail Latency:5.6
2024-11-13 16:58:35.290634617 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:58:35.290650522 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 8, QPS: 890, Avg Latency:8.98, Tail Latency:11.34
2024-11-13 16:58:37.784560529 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:58:37.784578336 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 16, QPS: 1168, Avg Latency:13.69, Tail Latency:15.7
2024-11-13 16:58:41.038908480 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:58:41.038926325 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 24, QPS: 1226, Avg Latency:19.57, Tail Latency:21.52
INFO:PerfEngine:Testing Finish. Report is saved in path: [ general_perf/reports/DCU/clip-onnx-fp32/result-fp32.json ]
INFO:PerfEngine:PDF Version is saved in path: [ general_perf/reports/DCU/clip-onnx-fp32/CLIP-ONNX-FP32-TO-FP32.JSON.pdf ]
ByteMLPerf/byte_infer_perf/general_perf/log/conformer-encoder-onnx-fp32.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:LANUCH:******************************************* Running CPU Numeric Checker... *******************************************
general_perf/backends/CPU/calculate_cpu_diff.sh: line 11: $'\r': command not found
INFO:CPUBase:Runing CPU Base...
INFO:BackendStore:Loading Compile Backend: CPU
INFO:BackendStore:Loading Runtime Backend: CPU
INFO:DatasetStore:Loading Dataset: Dataset does not exist, using fake data
INFO:FAKE_DATA:Rebatching batch size to: 4 ...
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/100 [00:00<?, ?it/s] 1%| | 1/100 [00:00<00:11, 8.88it/s] 2%|▏ | 2/100 [00:00<00:10, 9.45it/s] 3%|▎ | 3/100 [00:00<00:10, 9.59it/s] 5%|▌ | 5/100 [00:00<00:09, 9.83it/s] 7%|▋ | 7/100 [00:00<00:09, 9.94it/s] 9%|▉ | 9/100 [00:00<00:09, 10.01it/s] 10%|█ | 10/100 [00:01<00:09, 10.00it/s] 12%|█▏ | 12/100 [00:01<00:08, 10.02it/s] 14%|█▍ | 14/100 [00:01<00:08, 10.05it/s] 16%|█▌ | 16/100 [00:01<00:08, 10.07it/s] 18%|█▊ | 18/100 [00:01<00:08, 10.07it/s] 20%|██ | 20/100 [00:02<00:07, 10.07it/s] 22%|██▏ | 22/100 [00:02<00:07, 10.08it/s] 24%|██▍ | 24/100 [00:02<00:07, 10.09it/s] 26%|██▌ | 26/100 [00:02<00:07, 10.10it/s] 28%|██▊ | 28/100 [00:02<00:07, 10.10it/s] 30%|███ | 30/100 [00:02<00:06, 10.08it/s] 32%|███▏ | 32/100 [00:03<00:06, 10.07it/s] 34%|███▍ | 34/100 [00:03<00:06, 10.06it/s] 36%|███▌ | 36/100 [00:03<00:06, 10.06it/s] 38%|███▊ | 38/100 [00:03<00:06, 10.08it/s] 40%|████ | 40/100 [00:03<00:05, 10.08it/s] 42%|████▏ | 42/100 [00:04<00:05, 10.08it/s] 44%|████▍ | 44/100 [00:04<00:05, 10.08it/s] 46%|████▌ | 46/100 [00:04<00:05, 10.09it/s] 48%|████▊ | 48/100 [00:04<00:05, 10.10it/s] 50%|█████ | 50/100 [00:04<00:04, 10.08it/s] 52%|█████▏ | 52/100 [00:05<00:04, 10.08it/s] 54%|█████▍ | 54/100 [00:05<00:04, 10.08it/s] 56%|█████▌ | 56/100 [00:05<00:04, 10.09it/s] 58%|█████▊ | 58/100 [00:05<00:04, 10.10it/s] 60%|██████ | 60/100 [00:05<00:03, 10.09it/s] 62%|██████▏ | 62/100 [00:06<00:03, 10.12it/s] 64%|██████▍ | 64/100 [00:06<00:03, 10.09it/s] 66%|██████▌ | 66/100 [00:06<00:03, 10.11it/s] 68%|██████▊ | 68/100 [00:06<00:03, 10.11it/s] 70%|███████ | 70/100 [00:06<00:02, 10.12it/s] 72%|███████▏ | 72/100 [00:07<00:02, 10.11it/s] 74%|███████▍ | 74/100 [00:07<00:02, 10.08it/s] 76%|███████▌ | 76/100 [00:07<00:02, 10.09it/s] 78%|███████▊ | 78/100 [00:07<00:02, 10.10it/s] 80%|████████ | 80/100 [00:07<00:01, 10.11it/s] 82%|████████▏ | 82/100 [00:08<00:01, 10.10it/s] 84%|████████▍ | 84/100 [00:08<00:01, 10.11it/s] 86%|████████▌ | 86/100 [00:08<00:01, 10.09it/s] 88%|████████▊ | 88/100 [00:08<00:01, 10.10it/s] 90%|█████████ | 90/100 [00:08<00:00, 10.10it/s] 92%|█████████▏| 92/100 [00:09<00:00, 10.10it/s] 94%|█████████▍| 94/100 [00:09<00:00, 10.07it/s] 96%|█████████▌| 96/100 [00:09<00:00, 10.09it/s] 98%|█████████▊| 98/100 [00:09<00:00, 10.11it/s] 100%|██████████| 100/100 [00:09<00:00, 10.11it/s] 100%|██████████| 100/100 [00:09<00:00, 10.07it/s]
INFO:TestAccuracy:Batch size is 4, Accuracy: 0.0
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
INFO:PerfEngine:******************************************* Start to test model: conformer-encoder-onnx-fp32. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: fake_dataset
INFO:PerfEngine:Start to compile the model...
2024-11-13 16:38:21.047174738 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:38:21.047194536 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:PerfEngine:******************************************* Running Numeric Checker... *******************************************
INFO:FAKE_DATA:Rebatching batch size to: 4 ...
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/100 [00:00<?, ?it/s] 1%| | 1/100 [00:03<05:44, 3.48s/it] 6%|▌ | 6/100 [00:03<00:41, 2.24it/s] 11%|█ | 11/100 [00:03<00:18, 4.75it/s] 16%|█▌ | 16/100 [00:03<00:10, 7.87it/s] 20%|██ | 20/100 [00:03<00:07, 10.73it/s] 25%|██▌ | 25/100 [00:04<00:04, 15.09it/s] 30%|███ | 30/100 [00:04<00:03, 19.75it/s] 35%|███▌ | 35/100 [00:04<00:02, 24.38it/s] 40%|████ | 40/100 [00:04<00:02, 28.79it/s] 45%|████▌ | 45/100 [00:04<00:01, 32.69it/s] 50%|█████ | 50/100 [00:04<00:01, 36.03it/s] 55%|█████▌ | 55/100 [00:04<00:01, 38.71it/s] 60%|██████ | 60/100 [00:04<00:00, 40.80it/s] 65%|██████▌ | 65/100 [00:04<00:00, 42.33it/s] 70%|███████ | 70/100 [00:04<00:00, 43.48it/s] 75%|███████▌ | 75/100 [00:05<00:00, 44.33it/s] 80%|████████ | 80/100 [00:05<00:00, 44.92it/s] 85%|████████▌ | 85/100 [00:05<00:00, 45.38it/s] 90%|█████████ | 90/100 [00:05<00:00, 45.74it/s] 95%|█████████▌| 95/100 [00:05<00:00, 45.97it/s] 100%|██████████| 100/100 [00:05<00:00, 45.85it/s] 100%|██████████| 100/100 [00:05<00:00, 17.74it/s]
INFO:TestAccuracy:Batch size is 4, Accuracy: 0.0
INFO:TestAccuracy:Mean Diff: 4.478812343222671e-07, Std Diff: 4.088608136498806e-07, Max Diff: 1.52587890625e-05, Max Rel-Diff: 94.11724853515625, Mean Rel-Diff: 3.595714588300325e-05
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
2024-11-13 16:38:36.816556986 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:38:36.816574466 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 4, QPS: 394, Avg Latency:10.13, Tail Latency:12.36
2024-11-13 16:38:40.921813010 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:38:40.921831696 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 8, QPS: 462, Avg Latency:17.29, Tail Latency:19.83
2024-11-13 16:38:46.118517247 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:38:46.118534420 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 16, QPS: 499, Avg Latency:32.04, Tail Latency:34.71
2024-11-13 16:38:53.338765036 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:38:53.338781099 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 32, QPS: 894, Avg Latency:35.77, Tail Latency:37.95
2024-11-13 16:39:01.398073419 [W:onnxruntime:, session_state.cc:1169 VerifyEachNodeIsAssignedToAnEp] Some nodes were not assigned to the preferred execution providers which may or may not have an negative impact on performance. e.g. ORT explicitly assigns shape related ops to CPU to improve perf.
2024-11-13 16:39:01.398090794 [W:onnxruntime:, session_state.cc:1171 VerifyEachNodeIsAssignedToAnEp] Rerunning with verbose output on a non-minimal build will show node assignments.
INFO:BackendDCU:Batch size is 64, QPS: 942, Avg Latency:67.93, Tail Latency:70.5
INFO:PerfEngine:Testing Finish. Report is saved in path: [ general_perf/reports/DCU/conformer-encoder-onnx-fp32/result-fp32.json ]
ByteMLPerf/byte_infer_perf/general_perf/log/deberta-torch-fp32.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
INFO:PerfEngine:******************************************* Start to test model: deberta-torch-fp32. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_squad
INFO:SQUAD:Initial...
INFO:SQUAD:Preprocessing...
INFO:SQUAD:Rebatching batch size to: 11137 ...
0%| | 0/1 [00:00<?, ?it/s]100%|██████████| 1/1 [00:00<00:00, 95.79it/s]
INFO:PerfEngine:Start to compile the model...
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:SQUAD:Rebatching batch size to: 1 ...
0%| | 0/11137 [00:00<?, ?it/s] 66%|██████▌ | 7311/11137 [00:00<00:00, 18781.22it/s]100%|██████████| 11137/11137 [00:00<00:00, 27968.67it/s]
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/11137 [00:00<?, ?it/s] 0%| | 0/11137 [00:04<?, ?it/s]
Traceback (most recent call last):
File "/home/workspace/ByteMLPerf/byte_infer_perf/general_perf/core/perf_engine.py", line 412, in <module>
engine.start_engine()
File "/home/workspace/ByteMLPerf/byte_infer_perf/general_perf/core/perf_engine.py", line 107, in start_engine
status = self.single_workload_perf(self.workload)
File "/home/workspace/ByteMLPerf/byte_infer_perf/general_perf/core/perf_engine.py", line 212, in single_workload_perf
accuracy_results = AccuracyChecker.calculate_acc(
File "/home/workspace/ByteMLPerf/byte_infer_perf/general_perf/datasets/open_squad/test_accuracy.py", line 44, in calculate_acc
result = self.runtime_backend.predict(test_data)
File "/home/workspace/ByteMLPerf/byte_infer_perf/general_perf/backends/DCU/runtime_backend_dcu.py", line 96, in predict
results = model_runtime(*input_tensors)
File "/usr/local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript, serialized code (most recent call last):
File "code/__torch__.py", line 12, in forward
token_type_ids: Tensor) -> Tuple[Tensor, Tensor]:
model = self.model
_0 = (model).forward(input_ids, attention_mask, )
~~~~~~~~~~~~~~ <--- HERE
_1, _2, = _0
return (_1, _2)
File "code/__torch__/transformers/models/deberta/modeling_deberta.py", line 13, in forward
qa_outputs = self.qa_outputs
deberta = self.deberta
_0 = (deberta).forward(input_ids, attention_mask, )
~~~~~~~~~~~~~~~~ <--- HERE
_1 = torch.split((qa_outputs).forward(_0, ), 1, -1)
start_logits, end_logits, = _1
File "code/__torch__/transformers/models/deberta/modeling_deberta.py", line 32, in forward
embeddings = self.embeddings
_4 = (embeddings).forward(input_ids, attention_mask, )
_5 = (encoder).forward(attention_mask, _4, )
~~~~~~~~~~~~~~~~ <--- HERE
return _5
class DebertaEmbeddings(Module):
File "code/__torch__/transformers/models/deberta/modeling_deberta.py", line 138, in forward
rel_pos_ids0 = torch.slice(torch.slice(rel_pos_ids, 0, 0, _13), 1, 0, 9223372036854775807)
relative_pos = torch.unsqueeze(rel_pos_ids0, 0)
_20 = (_0).forward(weight, argument_2, relative_pos, mask, )
~~~~~~~~~~~ <--- HERE
_21 = (_1).forward(weight, _20, relative_pos, mask, )
_22 = (_2).forward(weight, _21, relative_pos, mask, )
File "code/__torch__/transformers/models/deberta/modeling_deberta.py", line 167, in forward
intermediate = self.intermediate
attention = self.attention
_32 = (attention).forward(rel_embeddings, argument_2, relative_pos, mask, )
~~~~~~~~~~~~~~~~~~ <--- HERE
_33 = (output).forward((intermediate).forward(_32, ), _32, )
return _33
File "code/__torch__/transformers/models/deberta/modeling_deberta.py", line 184, in forward
output = self.output
self0 = self.self
_34 = (self0).forward(rel_embeddings, argument_2, relative_pos, mask, )
~~~~~~~~~~~~~~ <--- HERE
return (output).forward(_34, argument_2, )
class DisentangledSelfAttention(Module):
File "code/__torch__/transformers/models/deberta/modeling_deberta.py", line 262, in forward
c2p_pos = torch.clamp(_63, 0, annotate(number, _64))
_65 = _35(c2p_pos, query_layer1, relative_pos2, )
c2p_att0 = torch.gather(c2p_att, -1, _65)
~~~~~~~~~~~~ <--- HERE
score = torch.add(c2p_att0, CONSTANTS.c5)
_66 = (pos_q_proj).forward(input, )
Traceback of TorchScript, original code (most recent call last):
/usr/local/lib/python3.7/dist-packages/transformers/models/deberta/modeling_deberta.py(666): disentangled_att_bias
/usr/local/lib/python3.7/dist-packages/transformers/models/deberta/modeling_deberta.py(617): forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1090): _slow_forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1102): _call_impl
/usr/local/lib/python3.7/dist-packages/transformers/models/deberta/modeling_deberta.py(271): forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1090): _slow_forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1102): _call_impl
/usr/local/lib/python3.7/dist-packages/transformers/models/deberta/modeling_deberta.py(338): forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1090): _slow_forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1102): _call_impl
/usr/local/lib/python3.7/dist-packages/transformers/models/deberta/modeling_deberta.py(432): forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1090): _slow_forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1102): _call_impl
/usr/local/lib/python3.7/dist-packages/transformers/models/deberta/modeling_deberta.py(937): forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1090): _slow_forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1102): _call_impl
/usr/local/lib/python3.7/dist-packages/transformers/models/deberta/modeling_deberta.py(1358): forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1090): _slow_forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1102): _call_impl
huggingface.py(80): forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1090): _slow_forward
/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py(1102): _call_impl
/usr/local/lib/python3.7/dist-packages/torch/jit/_trace.py(965): trace_module
/usr/local/lib/python3.7/dist-packages/torch/jit/_trace.py(750): trace
huggingface.py(87): <module>
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking argument for argument index in method wrapper_CUDA_gather)
ByteMLPerf/byte_infer_perf/general_perf/log/resnet50-onnxruntime-fp16.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
/usr/local/lib/python3.10/site-packages/tensorflow/python/keras/engine/training_arrays_v1.py:37: UserWarning: A NumPy version >=1.23.5 and <2.3.0 is required for this version of SciPy (detected version 1.23.0)
from scipy.sparse import issparse # pylint: disable=g-import-not-at-top
INFO:PerfEngine:******************************************* Start to test model: resnet50-onnxruntime-fp16. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_imagenet
INFO:Imagenet:Initial...
0it [00:00, ?it/s] 9297it [00:00, 92966.99it/s] 31650it [00:00, 169762.59it/s] 50000it [00:00, 177855.12it/s]
INFO:Imagenet:reduced image list, 45000 images not found
INFO:Imagenet:loaded 5000 images, cache=0, took=0.3sec
INFO:Imagenet:Preprocessing...
INFO:Imagenet:Rebatching batch size to: 1 ...
0%| | 0/5000 [00:00<?, ?it/s] 4%|▍ | 193/5000 [00:00<00:02, 1924.98it/s] 8%|▊ | 388/5000 [00:00<00:02, 1938.29it/s] 12%|█▏ | 582/5000 [00:00<00:02, 1919.19it/s] 16%|█▌ | 775/5000 [00:00<00:02, 1922.31it/s] 19%|█▉ | 968/5000 [00:00<00:02, 1921.08it/s] 23%|██▎ | 1161/5000 [00:00<00:02, 1916.49it/s] 27%|██▋ | 1357/5000 [00:00<00:01, 1929.47it/s] 31%|███ | 1555/5000 [00:00<00:01, 1942.96it/s] 35%|███▌ | 1753/5000 [00:00<00:01, 1953.38it/s] 39%|███▉ | 1951/5000 [00:01<00:01, 1960.21it/s] 43%|████▎ | 2148/5000 [00:01<00:01, 1937.94it/s] 47%|████▋ | 2346/5000 [00:01<00:01, 1948.89it/s] 51%|█████ | 2541/5000 [00:01<00:01, 1945.09it/s] 55%|█████▍ | 2736/5000 [00:01<00:01, 1934.50it/s] 59%|█████▊ | 2933/5000 [00:01<00:01, 1943.05it/s] 63%|██████▎ | 3129/5000 [00:01<00:00, 1945.57it/s] 67%|██████▋ | 3326/5000 [00:01<00:00, 1950.60it/s] 70%|███████ | 3523/5000 [00:01<00:00, 1953.93it/s] 74%|███████▍ | 3719/5000 [00:01<00:00, 1955.51it/s] 78%|███████▊ | 3915/5000 [00:02<00:00, 1944.63it/s] 82%|████████▏ | 4112/5000 [00:02<00:00, 1949.95it/s] 86%|████████▌ | 4308/5000 [00:02<00:00, 1947.67it/s] 90%|█████████ | 4506/5000 [00:02<00:00, 1954.55it/s] 94%|█████████▍| 4702/5000 [00:02<00:00, 1950.45it/s] 98%|█████████▊| 4899/5000 [00:02<00:00, 1955.93it/s] 100%|██████████| 5000/5000 [00:02<00:00, 1944.71it/s]
INFO:PerfEngine:Start to compile the model...
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:Imagenet:Rebatching batch size to: 1 ...
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/5000 [00:00<?, ?it/s] 0%| | 1/5000 [00:03<4:20:34, 3.13s/it] 1%| | 32/5000 [00:03<05:59, 13.84it/s] 1%|▏ | 63/5000 [00:03<02:38, 31.22it/s] 2%|▏ | 94/5000 [00:03<01:32, 52.77it/s] 2%|▎ | 125/5000 [00:03<01:02, 78.35it/s] 3%|▎ | 156/5000 [00:03<00:45, 107.15it/s] 4%|▎ | 187/5000 [00:03<00:35, 137.34it/s] 4%|▍ | 218/5000 [00:03<00:28, 167.80it/s] 5%|▍ | 249/5000 [00:03<00:24, 196.33it/s] 6%|▌ | 280/5000 [00:04<00:21, 220.36it/s] 6%|▌ | 311/5000 [00:04<00:19, 241.76it/s] 7%|▋ | 342/5000 [00:04<00:17, 259.01it/s] 7%|▋ | 373/5000 [00:04<00:16, 272.36it/s] 8%|▊ | 404/5000 [00:04<00:16, 282.58it/s] 9%|▊ | 435/5000 [00:04<00:15, 289.95it/s] 9%|▉ | 466/5000 [00:04<00:15, 295.51it/s] 10%|▉ | 497/5000 [00:04<00:15, 299.45it/s] 11%|█ | 528/5000 [00:04<00:14, 302.40it/s] 11%|█ | 559/5000 [00:04<00:14, 304.45it/s] 12%|█▏ | 590/5000 [00:05<00:14, 306.08it/s] 12%|█▏ | 621/5000 [00:05<00:14, 306.97it/s] 13%|█▎ | 652/5000 [00:05<00:14, 307.69it/s] 14%|█▎ | 683/5000 [00:05<00:14, 306.11it/s] 14%|█▍ | 714/5000 [00:05<00:13, 306.89it/s] 15%|█▍ | 745/5000 [00:05<00:13, 307.44it/s] 16%|█▌ | 776/5000 [00:05<00:13, 307.91it/s] 16%|█▌ | 807/5000 [00:05<00:13, 308.34it/s] 17%|█▋ | 838/5000 [00:05<00:13, 308.44it/s] 17%|█▋ | 869/5000 [00:05<00:13, 308.67it/s] 18%|█▊ | 900/5000 [00:06<00:13, 308.70it/s] 19%|█▊ | 931/5000 [00:06<00:13, 308.85it/s] 19%|█▉ | 963/5000 [00:06<00:13, 309.19it/s] 20%|█▉ | 994/5000 [00:06<00:12, 309.23it/s] 20%|██ | 1025/5000 [00:06<00:12, 309.36it/s] 21%|██ | 1056/5000 [00:06<00:12, 309.54it/s] 22%|██▏ | 1087/5000 [00:06<00:12, 307.45it/s] 22%|██▏ | 1118/5000 [00:06<00:12, 308.13it/s] 23%|██▎ | 1149/5000 [00:06<00:12, 308.50it/s] 24%|██▎ | 1181/5000 [00:06<00:12, 309.04it/s] 24%|██▍ | 1213/5000 [00:07<00:12, 309.39it/s] 25%|██▍ | 1244/5000 [00:07<00:12, 309.53it/s] 26%|██▌ | 1275/5000 [00:07<00:12, 309.62it/s] 26%|██▌ | 1307/5000 [00:07<00:11, 309.74it/s] 27%|██▋ | 1338/5000 [00:07<00:11, 309.61it/s] 27%|██▋ | 1369/5000 [00:07<00:11, 309.69it/s] 28%|██▊ | 1400/5000 [00:07<00:11, 309.73it/s] 29%|██▊ | 1432/5000 [00:07<00:11, 309.87it/s] 29%|██▉ | 1463/5000 [00:07<00:11, 309.86it/s] 30%|██▉ | 1494/5000 [00:07<00:11, 309.70it/s] 30%|███ | 1525/5000 [00:08<00:11, 307.48it/s] 31%|███ | 1556/5000 [00:08<00:11, 308.10it/s] 32%|███▏ | 1587/5000 [00:08<00:11, 308.55it/s] 32%|███▏ | 1618/5000 [00:08<00:10, 308.95it/s] 33%|███▎ | 1649/5000 [00:08<00:10, 309.12it/s] 34%|███▎ | 1680/5000 [00:08<00:10, 309.13it/s] 34%|███▍ | 1711/5000 [00:08<00:10, 309.26it/s] 35%|███▍ | 1742/5000 [00:08<00:10, 308.93it/s] 35%|███▌ | 1773/5000 [00:08<00:10, 309.02it/s] 36%|███▌ | 1804/5000 [00:08<00:10, 309.23it/s] 37%|███▋ | 1835/5000 [00:09<00:10, 309.22it/s] 37%|███▋ | 1866/5000 [00:09<00:10, 309.41it/s] 38%|███▊ | 1897/5000 [00:09<00:10, 309.41it/s] 39%|███▊ | 1928/5000 [00:09<00:10, 307.00it/s] 39%|███▉ | 1959/5000 [00:09<00:09, 307.79it/s] 40%|███▉ | 1990/5000 [00:09<00:09, 308.18it/s] 40%|████ | 2021/5000 [00:09<00:09, 308.48it/s] 41%|████ | 2052/5000 [00:09<00:09, 308.74it/s] 42%|████▏ | 2083/5000 [00:09<00:09, 308.75it/s] 42%|████▏ | 2114/5000 [00:09<00:09, 308.91it/s] 43%|████▎ | 2146/5000 [00:10<00:09, 309.24it/s] 44%|████▎ | 2177/5000 [00:10<00:09, 308.92it/s] 44%|████▍ | 2208/5000 [00:10<00:09, 307.90it/s] 45%|████▍ | 2239/5000 [00:10<00:08, 308.33it/s] 45%|████▌ | 2270/5000 [00:10<00:08, 308.52it/s] 46%|████▌ | 2301/5000 [00:10<00:08, 308.84it/s] 47%|████▋ | 2332/5000 [00:10<00:08, 306.61it/s] 47%|████▋ | 2363/5000 [00:10<00:08, 307.22it/s] 48%|████▊ | 2394/5000 [00:10<00:08, 307.90it/s] 48%|████▊ | 2425/5000 [00:10<00:08, 308.31it/s] 49%|████▉ | 2456/5000 [00:11<00:08, 308.52it/s] 50%|████▉ | 2487/5000 [00:11<00:08, 308.53it/s] 50%|█████ | 2518/5000 [00:11<00:08, 308.52it/s] 51%|█████ | 2549/5000 [00:11<00:07, 308.82it/s] 52%|█████▏ | 2580/5000 [00:11<00:07, 309.11it/s] 52%|█████▏ | 2611/5000 [00:11<00:07, 309.17it/s] 53%|█████▎ | 2642/5000 [00:11<00:07, 309.28it/s] 53%|█████▎ | 2673/5000 [00:11<00:07, 309.05it/s] 54%|█████▍ | 2704/5000 [00:11<00:07, 308.93it/s] 55%|█████▍ | 2735/5000 [00:11<00:07, 309.02it/s] 55%|█████▌ | 2766/5000 [00:12<00:07, 307.03it/s] 56%|█████▌ | 2797/5000 [00:12<00:07, 307.68it/s] 57%|█████▋ | 2828/5000 [00:12<00:07, 308.27it/s] 57%|█████▋ | 2859/5000 [00:12<00:06, 308.59it/s] 58%|█████▊ | 2890/5000 [00:12<00:06, 308.66it/s] 58%|█████▊ | 2921/5000 [00:12<00:06, 308.82it/s] 59%|█████▉ | 2952/5000 [00:12<00:06, 309.11it/s] 60%|█████▉ | 2983/5000 [00:12<00:06, 309.19it/s] 60%|██████ | 3014/5000 [00:12<00:06, 309.21it/s] 61%|██████ | 3045/5000 [00:12<00:06, 309.12it/s] 62%|██████▏ | 3076/5000 [00:13<00:06, 309.25it/s] 62%|██████▏ | 3107/5000 [00:13<00:06, 309.13it/s] 63%|██████▎ | 3138/5000 [00:13<00:06, 309.08it/s] 63%|██████▎ | 3169/5000 [00:13<00:05, 307.63it/s] 64%|██████▍ | 3200/5000 [00:13<00:05, 307.91it/s] 65%|██████▍ | 3231/5000 [00:13<00:05, 308.30it/s] 65%|██████▌ | 3262/5000 [00:13<00:05, 308.55it/s] 66%|██████▌ | 3293/5000 [00:13<00:05, 308.59it/s] 66%|██████▋ | 3324/5000 [00:13<00:05, 308.48it/s] 67%|██████▋ | 3355/5000 [00:13<00:05, 308.82it/s] 68%|██████▊ | 3386/5000 [00:14<00:05, 308.85it/s] 68%|██████▊ | 3417/5000 [00:14<00:05, 308.55it/s] 69%|██████▉ | 3448/5000 [00:14<00:05, 308.76it/s] 70%|██████▉ | 3479/5000 [00:14<00:04, 308.75it/s] 70%|███████ | 3510/5000 [00:14<00:04, 308.82it/s] 71%|███████ | 3541/5000 [00:14<00:04, 308.87it/s] 71%|███████▏ | 3572/5000 [00:14<00:04, 306.68it/s] 72%|███████▏ | 3603/5000 [00:14<00:04, 307.56it/s] 73%|███████▎ | 3634/5000 [00:14<00:04, 308.08it/s] 73%|███████▎ | 3665/5000 [00:15<00:04, 308.37it/s] 74%|███████▍ | 3696/5000 [00:15<00:04, 308.71it/s] 75%|███████▍ | 3727/5000 [00:15<00:04, 308.81it/s] 75%|███████▌ | 3758/5000 [00:15<00:04, 308.83it/s] 76%|███████▌ | 3789/5000 [00:15<00:03, 308.78it/s] 76%|███████▋ | 3820/5000 [00:15<00:03, 308.94it/s] 77%|███████▋ | 3851/5000 [00:15<00:03, 309.03it/s] 78%|███████▊ | 3882/5000 [00:15<00:03, 308.38it/s] 78%|███████▊ | 3913/5000 [00:15<00:03, 308.53it/s] 79%|███████▉ | 3944/5000 [00:15<00:03, 308.86it/s] 80%|███████▉ | 3975/5000 [00:16<00:03, 308.78it/s] 80%|████████ | 4006/5000 [00:16<00:03, 306.48it/s] 81%|████████ | 4037/5000 [00:16<00:03, 307.35it/s] 81%|████████▏ | 4068/5000 [00:16<00:03, 307.86it/s] 82%|████████▏ | 4099/5000 [00:16<00:02, 308.16it/s] 83%|████████▎ | 4130/5000 [00:16<00:02, 308.57it/s] 83%|████████▎ | 4161/5000 [00:16<00:02, 308.62it/s] 84%|████████▍ | 4192/5000 [00:16<00:02, 308.83it/s] 84%|████████▍ | 4223/5000 [00:16<00:02, 309.13it/s] 85%|████████▌ | 4254/5000 [00:16<00:02, 309.04it/s] 86%|████████▌ | 4285/5000 [00:17<00:02, 309.01it/s] 86%|████████▋ | 4316/5000 [00:17<00:02, 309.17it/s] 87%|████████▋ | 4347/5000 [00:17<00:02, 309.18it/s] 88%|████████▊ | 4378/5000 [00:17<00:02, 309.38it/s] 88%|████████▊ | 4409/5000 [00:17<00:01, 306.77it/s] 89%|████████▉ | 4440/5000 [00:17<00:01, 307.39it/s] 89%|████████▉ | 4471/5000 [00:17<00:01, 307.95it/s] 90%|█████████ | 4502/5000 [00:17<00:01, 308.31it/s] 91%|█████████ | 4533/5000 [00:17<00:01, 308.65it/s] 91%|█████████▏| 4564/5000 [00:17<00:01, 308.70it/s] 92%|█████████▏| 4595/5000 [00:18<00:01, 308.77it/s] 93%|█████████▎| 4627/5000 [00:18<00:01, 309.21it/s] 93%|█████████▎| 4658/5000 [00:18<00:01, 309.16it/s] 94%|█████████▍| 4689/5000 [00:18<00:01, 309.08it/s] 94%|█████████▍| 4720/5000 [00:18<00:00, 309.00it/s] 95%|█████████▌| 4751/5000 [00:18<00:00, 308.87it/s] 96%|█████████▌| 4782/5000 [00:18<00:00, 308.29it/s] 96%|█████████▋| 4813/5000 [00:18<00:00, 306.30it/s] 97%|█████████▋| 4844/5000 [00:18<00:00, 307.38it/s] 98%|█████████▊| 4875/5000 [00:18<00:00, 307.91it/s] 98%|█████████▊| 4906/5000 [00:19<00:00, 308.05it/s] 99%|█████████▊| 4937/5000 [00:19<00:00, 308.37it/s] 99%|█████████▉| 4968/5000 [00:19<00:00, 308.58it/s] 100%|█████████▉| 4999/5000 [00:19<00:00, 308.92it/s] 100%|██████████| 5000/5000 [00:19<00:00, 258.69it/s]
INFO:TestAccuracy:Batch size is 1, Accuracy: 0.7698
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
INFO:BackendDCU:Batch size is 1, QPS: 373, Avg Latency:2.68, Tail Latency:5.21
INFO:BackendDCU:Batch size is 2, QPS: 586, Avg Latency:3.41, Tail Latency:3.43
INFO:BackendDCU:Batch size is 4, QPS: 867, Avg Latency:4.61, Tail Latency:4.72
INFO:BackendDCU:Batch size is 8, QPS: 1271, Avg Latency:6.29, Tail Latency:6.45
INFO:BackendDCU:Batch size is 16, QPS: 1729, Avg Latency:9.25, Tail Latency:11.55
INFO:BackendDCU:Batch size is 32, QPS: 2047, Avg Latency:15.63, Tail Latency:17.64
INFO:BackendDCU:Batch size is 64, QPS: 2088, Avg Latency:30.65, Tail Latency:33.75
INFO:BackendDCU:Batch size is 128, QPS: 2084, Avg Latency:61.41, Tail Latency:63.78
INFO:BackendDCU:Batch size is 256, QPS: 1996, Avg Latency:128.23, Tail Latency:131.11
INFO:BackendDCU:Batch size is 512, QPS: 1955, Avg Latency:261.82, Tail Latency:264.08
INFO:BackendDCU:Batch size is 1024, QPS: 1883, Avg Latency:543.67, Tail Latency:546.07
INFO:PerfEngine:Testing Finish. Report is saved in path: [ general_perf/reports/DCU/resnet50-onnxruntime-fp16/result-fp16.json ]
INFO:PerfEngine:PDF Version is saved in path: [ general_perf/reports/DCU/resnet50-onnxruntime-fp16/RESNET50-ONNXRUNTIME-FP16-TO-FP16.JSON.pdf ]
ByteMLPerf/byte_infer_perf/general_perf/log/resnet50-onnxruntime-fp32.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
/usr/local/lib/python3.10/site-packages/tensorflow/python/keras/engine/training_arrays_v1.py:37: UserWarning: A NumPy version >=1.23.5 and <2.3.0 is required for this version of SciPy (detected version 1.23.0)
from scipy.sparse import issparse # pylint: disable=g-import-not-at-top
INFO:PerfEngine:******************************************* Start to test model: resnet50-onnxruntime-fp32. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_imagenet
INFO:Imagenet:Initial...
0it [00:00, ?it/s] 8954it [00:00, 89533.04it/s] 31192it [00:00, 167669.79it/s] 50000it [00:00, 176304.75it/s]
INFO:Imagenet:reduced image list, 45000 images not found
INFO:Imagenet:loaded 5000 images, cache=0, took=0.3sec
INFO:Imagenet:Preprocessing...
INFO:Imagenet:Rebatching batch size to: 1 ...
0%| | 0/5000 [00:00<?, ?it/s] 4%|▎ | 186/5000 [00:00<00:02, 1855.51it/s] 8%|▊ | 375/5000 [00:00<00:02, 1872.24it/s] 11%|█▏ | 564/5000 [00:00<00:02, 1878.52it/s] 15%|█▌ | 752/5000 [00:00<00:02, 1875.77it/s] 19%|█▉ | 942/5000 [00:00<00:02, 1881.55it/s] 23%|██▎ | 1131/5000 [00:00<00:02, 1884.18it/s] 26%|██▋ | 1321/5000 [00:00<00:01, 1888.55it/s] 30%|███ | 1513/5000 [00:00<00:01, 1897.76it/s] 34%|███▍ | 1704/5000 [00:00<00:01, 1901.48it/s] 38%|███▊ | 1896/5000 [00:01<00:01, 1904.10it/s] 42%|████▏ | 2087/5000 [00:01<00:01, 1898.99it/s] 46%|████▌ | 2279/5000 [00:01<00:01, 1904.55it/s] 49%|████▉ | 2470/5000 [00:01<00:01, 1904.08it/s] 53%|█████▎ | 2661/5000 [00:01<00:01, 1902.00it/s] 57%|█████▋ | 2852/5000 [00:01<00:01, 1900.51it/s] 61%|██████ | 3044/5000 [00:01<00:01, 1906.20it/s] 65%|██████▍ | 3236/5000 [00:01<00:00, 1908.16it/s] 69%|██████▊ | 3429/5000 [00:01<00:00, 1912.50it/s] 72%|███████▏ | 3621/5000 [00:01<00:00, 1911.03it/s] 76%|███████▋ | 3813/5000 [00:02<00:00, 1904.16it/s] 80%|████████ | 4006/5000 [00:02<00:00, 1908.76it/s] 84%|████████▍ | 4197/5000 [00:02<00:00, 1907.56it/s] 88%|████████▊ | 4388/5000 [00:02<00:00, 1904.84it/s] 92%|█████████▏| 4581/5000 [00:02<00:00, 1909.25it/s] 95%|█████████▌| 4773/5000 [00:02<00:00, 1912.28it/s] 99%|█████████▉| 4966/5000 [00:02<00:00, 1916.50it/s] 100%|██████████| 5000/5000 [00:02<00:00, 1901.84it/s]
INFO:PerfEngine:Start to compile the model...
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:Imagenet:Rebatching batch size to: 1 ...
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/5000 [00:00<?, ?it/s] 0%| | 1/5000 [00:03<4:24:22, 3.17s/it] 1%| | 30/5000 [00:03<06:28, 12.78it/s] 1%| | 59/5000 [00:03<02:51, 28.81it/s] 2%|▏ | 88/5000 [00:03<01:40, 48.71it/s] 2%|▏ | 117/5000 [00:03<01:07, 72.33it/s] 3%|▎ | 146/5000 [00:03<00:49, 98.95it/s] 4%|▎ | 175/5000 [00:03<00:37, 127.26it/s] 4%|▍ | 204/5000 [00:03<00:30, 155.51it/s] 5%|▍ | 233/5000 [00:03<00:26, 181.99it/s] 5%|▌ | 262/5000 [00:04<00:23, 205.47it/s] 6%|▌ | 291/5000 [00:04<00:20, 225.18it/s] 6%|▋ | 320/5000 [00:04<00:19, 239.56it/s] 7%|▋ | 349/5000 [00:04<00:18, 252.17it/s] 8%|▊ | 378/5000 [00:04<00:17, 261.70it/s] 8%|▊ | 407/5000 [00:04<00:17, 268.72it/s] 9%|▊ | 436/5000 [00:04<00:16, 273.88it/s] 9%|▉ | 465/5000 [00:04<00:16, 277.68it/s]
\ No newline at end of file
ByteMLPerf/byte_infer_perf/general_perf/log/resnet50-tf-fp32.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
INFO:PerfEngine:******************************************* Start to test model: resnet50-tf-fp32. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_imagenet
INFO:Imagenet:Initial...
0it [00:00, ?it/s] 9265it [00:00, 92645.01it/s] 31721it [00:00, 170236.62it/s] 50000it [00:00, 178218.93it/s]
INFO:Imagenet:reduced image list, 45000 images not found
INFO:Imagenet:loaded 5000 images, cache=0, took=0.3sec
INFO:Imagenet:Preprocessing...
INFO:Imagenet:Rebatching batch size to: 1 ...
0%| | 0/5000 [00:00<?, ?it/s] 4%|▎ | 177/5000 [00:00<00:02, 1767.72it/s] 7%|▋ | 355/5000 [00:00<00:02, 1771.08it/s] 11%|█ | 534/5000 [00:00<00:02, 1779.36it/s] 14%|█▍ | 716/5000 [00:00<00:02, 1794.49it/s] 18%|█▊ | 896/5000 [00:00<00:02, 1794.49it/s] 22%|██▏ | 1080/5000 [00:00<00:02, 1809.24it/s] 25%|██▌ | 1264/5000 [00:00<00:02, 1817.38it/s] 29%|██▉ | 1447/5000 [00:00<00:01, 1817.85it/s] 33%|███▎ | 1633/5000 [00:00<00:01, 1829.06it/s] 36%|███▋ | 1818/5000 [00:01<00:01, 1835.16it/s] 40%|████ | 2003/5000 [00:01<00:01, 1837.14it/s] 44%|████▎ | 2187/5000 [00:01<00:01, 1833.66it/s] 47%|████▋ | 2371/5000 [00:01<00:01, 1833.66it/s] 51%|█████ | 2556/5000 [00:01<00:01, 1837.28it/s] 55%|█████▍ | 2740/5000 [00:01<00:01, 1834.83it/s] 58%|█████▊ | 2924/5000 [00:01<00:01, 1835.65it/s] 62%|██████▏ | 3112/5000 [00:01<00:01, 1846.61it/s] 66%|██████▌ | 3297/5000 [00:01<00:00, 1847.08it/s] 70%|██████▉ | 3482/5000 [00:01<00:00, 1843.11it/s] 73%|███████▎ | 3668/5000 [00:02<00:00, 1846.73it/s] 77%|███████▋ | 3855/5000 [00:02<00:00, 1851.02it/s] 81%|████████ | 4041/5000 [00:02<00:00, 1852.22it/s] 85%|████████▍ | 4227/5000 [00:02<00:00, 1849.42it/s] 88%|████████▊ | 4414/5000 [00:02<00:00, 1853.42it/s] 92%|█████████▏| 4600/5000 [00:02<00:00, 1852.90it/s] 96%|█████████▌| 4786/5000 [00:02<00:00, 1847.07it/s] 99%|█████████▉| 4974/5000 [00:02<00:00, 1854.41it/s] 100%|██████████| 5000/5000 [00:02<00:00, 1834.66it/s]
INFO:PerfEngine:Start to compile the model...
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:absl:Fingerprint not found. Saved model loading will continue.
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:Imagenet:Rebatching batch size to: 4 ...
0%| | 0/1250 [00:00<?, ?it/s] 5%|▍ | 62/1250 [00:00<00:01, 611.78it/s] 10%|▉ | 124/1250 [00:00<00:02, 507.21it/s] 14%|█▍ | 176/1250 [00:00<00:02, 478.38it/s] 18%|█▊ | 225/1250 [00:00<00:02, 465.93it/s] 22%|██▏ | 272/1250 [00:00<00:02, 461.66it/s] 26%|██▌ | 319/1250 [00:00<00:02, 459.86it/s] 29%|██▉ | 366/1250 [00:00<00:01, 457.09it/s] 33%|███▎ | 412/1250 [00:00<00:01, 454.87it/s] 37%|███▋ | 458/1250 [00:00<00:01, 452.81it/s] 40%|████ | 504/1250 [00:01<00:01, 451.72it/s] 44%|████▍ | 550/1250 [00:01<00:01, 451.50it/s] 48%|████▊ | 596/1250 [00:01<00:01, 451.14it/s] 51%|█████▏ | 642/1250 [00:01<00:01, 450.97it/s] 55%|█████▌ | 688/1250 [00:01<00:01, 450.63it/s] 59%|█████▊ | 734/1250 [00:01<00:01, 450.70it/s] 62%|██████▏ | 780/1250 [00:01<00:01, 450.69it/s] 66%|██████▌ | 826/1250 [00:01<00:00, 451.79it/s] 70%|██████▉ | 872/1250 [00:01<00:00, 453.23it/s] 73%|███████▎ | 918/1250 [00:02<00:00, 452.97it/s] 77%|███████▋ | 964/1250 [00:02<00:00, 452.63it/s] 81%|████████ | 1010/1250 [00:02<00:00, 451.31it/s] 84%|████████▍ | 1056/1250 [00:02<00:00, 450.84it/s] 88%|████████▊ | 1102/1250 [00:02<00:00, 451.02it/s] 92%|█████████▏| 1148/1250 [00:02<00:00, 450.13it/s] 96%|█████████▌| 1194/1250 [00:02<00:00, 450.20it/s] 99%|█████████▉| 1240/1250 [00:02<00:00, 450.28it/s] 100%|██████████| 1250/1250 [00:02<00:00, 456.54it/s]
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/1250 [00:00<?, ?it/s]WARNING:absl:`input_tensor:0` is not a valid tf.function parameter name. Sanitizing to `input_tensor_0`.
0%| | 1/1250 [00:08<2:47:44, 8.06s/it]WARNING:absl:`input_tensor:0` is not a valid tf.function parameter name. Sanitizing to `input_tensor_0`.
WARNING:absl:`input_tensor:0` is not a valid tf.function parameter name. Sanitizing to `input_tensor_0`.
WARNING:absl:`input_tensor:0` is not a valid tf.function parameter name. Sanitizing to `input_tensor_0`.
WARNING:absl:`input_tensor:0` is not a valid tf.function parameter name. Sanitizing to `input_tensor_0`.
1%| | 12/1250 [00:08<10:07, 2.04it/s] 2%|▏ | 23/1250 [00:08<04:23, 4.65it/s] 3%|▎ | 33/1250 [00:08<02:35, 7.84it/s] 3%|▎ | 43/1250 [00:08<01:40, 12.01it/s] 4%|▍ | 53/1250 [00:08<01:09, 17.31it/s] 5%|▌ | 63/1250 [00:08<00:49, 23.77it/s] 6%|▌ | 73/1250 [00:08<00:37, 31.28it/s] 7%|▋ | 83/1250 [00:08<00:29, 39.55it/s] 7%|▋ | 93/1250 [00:09<00:24, 48.09it/s] 8%|▊ | 103/1250 [00:09<00:20, 56.40it/s] 9%|▉ | 113/1250 [00:09<00:17, 64.01it/s] 10%|▉ | 123/1250 [00:09<00:16, 70.41it/s] 11%|█ | 133/1250 [00:09<00:14, 76.15it/s] 11%|█▏ | 143/1250 [00:09<00:13, 80.78it/s] 12%|█▏ | 153/1250 [00:09<00:13, 84.25it/s] 13%|█▎ | 163/1250 [00:09<00:12, 86.89it/s] 14%|█▍ | 173/1250 [00:09<00:12, 88.94it/s] 15%|█▍ | 183/1250 [00:09<00:11, 90.34it/s] 15%|█▌ | 193/1250 [00:10<00:11, 91.44it/s] 16%|█▌ | 203/1250 [00:10<00:11, 92.18it/s] 17%|█▋ | 213/1250 [00:10<00:11, 92.46it/s] 18%|█▊ | 223/1250 [00:10<00:11, 92.91it/s] 19%|█▊ | 233/1250 [00:10<00:10, 93.21it/s] 19%|█▉ | 243/1250 [00:10<00:10, 92.96it/s] 20%|██ | 253/1250 [00:10<00:10, 93.32it/s] 21%|██ | 263/1250 [00:10<00:10, 93.54it/s] 22%|██▏ | 273/1250 [00:10<00:10, 93.60it/s] 23%|██▎ | 283/1250 [00:11<00:10, 93.76it/s] 23%|██▎ | 293/1250 [00:11<00:10, 93.72it/s] 24%|██▍ | 303/1250 [00:11<00:10, 93.67it/s] 25%|██▌ | 314/1250 [00:11<00:09, 97.14it/s] 26%|██▌ | 324/1250 [00:11<00:09, 96.99it/s] 27%|██▋ | 334/1250 [00:11<00:09, 96.10it/s] 28%|██▊ | 344/1250 [00:11<00:09, 95.57it/s] 28%|██▊ | 354/1250 [00:11<00:09, 95.17it/s] 29%|██▉ | 364/1250 [00:11<00:09, 94.76it/s] 30%|██▉ | 374/1250 [00:11<00:09, 94.06it/s] 31%|███ | 384/1250 [00:12<00:09, 94.04it/s] 32%|███▏ | 394/1250 [00:12<00:09, 93.99it/s] 32%|███▏ | 404/1250 [00:12<00:09, 93.84it/s] 33%|███▎ | 414/1250 [00:12<00:08, 93.97it/s] 34%|███▍ | 424/1250 [00:12<00:08, 93.91it/s] 35%|███▍ | 434/1250 [00:12<00:08, 93.83it/s] 36%|███▌ | 444/1250 [00:12<00:08, 93.98it/s] 36%|███▋ | 454/1250 [00:12<00:08, 93.84it/s] 37%|███▋ | 464/1250 [00:12<00:08, 93.94it/s] 38%|███▊ | 474/1250 [00:13<00:08, 93.90it/s] 39%|███▊ | 484/1250 [00:13<00:08, 93.92it/s] 40%|███▉ | 494/1250 [00:13<00:08, 93.33it/s] 40%|████ | 504/1250 [00:13<00:07, 93.44it/s] 41%|████ | 514/1250 [00:13<00:07, 93.58it/s] 42%|████▏ | 524/1250 [00:13<00:07, 93.76it/s] 43%|████▎ | 534/1250 [00:13<00:07, 93.86it/s] 44%|████▎ | 544/1250 [00:13<00:07, 93.86it/s] 44%|████▍ | 554/1250 [00:13<00:07, 93.89it/s] 45%|████▌ | 564/1250 [00:14<00:07, 93.87it/s] 46%|████▌ | 574/1250 [00:14<00:07, 93.82it/s] 47%|████▋ | 584/1250 [00:14<00:07, 93.76it/s] 48%|████▊ | 594/1250 [00:14<00:07, 93.64it/s] 48%|████▊ | 604/1250 [00:14<00:06, 93.63it/s] 49%|████▉ | 614/1250 [00:14<00:06, 93.83it/s] 50%|████▉ | 624/1250 [00:14<00:06, 93.00it/s] 51%|█████ | 634/1250 [00:14<00:06, 93.33it/s] 52%|█████▏ | 644/1250 [00:14<00:06, 93.56it/s] 52%|█████▏ | 654/1250 [00:14<00:06, 93.73it/s] 53%|█████▎ | 664/1250 [00:15<00:06, 93.80it/s] 54%|█████▍ | 674/1250 [00:15<00:06, 93.81it/s] 55%|█████▍ | 684/1250 [00:15<00:06, 93.64it/s] 56%|█████▌ | 694/1250 [00:15<00:05, 93.66it/s] 56%|█████▋ | 704/1250 [00:15<00:05, 93.64it/s] 57%|█████▋ | 714/1250 [00:15<00:05, 93.78it/s] 58%|█████▊ | 724/1250 [00:15<00:05, 93.66it/s] 59%|█████▊ | 734/1250 [00:15<00:05, 93.78it/s] 60%|█████▉ | 744/1250 [00:15<00:05, 93.21it/s] 60%|██████ | 754/1250 [00:16<00:05, 93.43it/s] 61%|██████ | 764/1250 [00:16<00:05, 93.54it/s] 62%|██████▏ | 774/1250 [00:16<00:05, 93.44it/s] 63%|██████▎ | 784/1250 [00:16<00:04, 93.64it/s] 64%|██████▎ | 794/1250 [00:16<00:04, 93.85it/s] 64%|██████▍ | 804/1250 [00:16<00:04, 93.87it/s] 65%|██████▌ | 814/1250 [00:16<00:04, 94.00it/s] 66%|██████▌ | 824/1250 [00:16<00:04, 93.96it/s] 67%|██████▋ | 834/1250 [00:16<00:04, 93.97it/s] 68%|██████▊ | 844/1250 [00:17<00:04, 93.86it/s] 68%|██████▊ | 854/1250 [00:17<00:04, 93.85it/s] 69%|██████▉ | 864/1250 [00:17<00:04, 93.73it/s] 70%|██████▉ | 874/1250 [00:17<00:04, 93.14it/s] 71%|███████ | 884/1250 [00:17<00:03, 93.29it/s] 72%|███████▏ | 894/1250 [00:17<00:03, 93.36it/s] 72%|███████▏ | 904/1250 [00:17<00:03, 93.44it/s] 73%|███████▎ | 914/1250 [00:17<00:03, 93.69it/s] 74%|███████▍ | 924/1250 [00:17<00:03, 93.89it/s] 75%|███████▍ | 934/1250 [00:17<00:03, 94.01it/s] 76%|███████▌ | 944/1250 [00:18<00:03, 94.03it/s] 76%|███████▋ | 954/1250 [00:18<00:03, 94.05it/s] 77%|███████▋ | 964/1250 [00:18<00:03, 94.01it/s] 78%|███████▊ | 974/1250 [00:18<00:02, 93.95it/s] 79%|███████▊ | 984/1250 [00:18<00:02, 93.93it/s] 80%|███████▉ | 994/1250 [00:18<00:02, 93.80it/s] 80%|████████ | 1004/1250 [00:18<00:02, 93.41it/s] 81%|████████ | 1014/1250 [00:18<00:02, 93.64it/s] 82%|████████▏ | 1024/1250 [00:18<00:02, 93.62it/s] 83%|████████▎ | 1034/1250 [00:19<00:02, 93.82it/s] 84%|████████▎ | 1044/1250 [00:19<00:02, 93.96it/s] 84%|████████▍ | 1054/1250 [00:19<00:02, 93.99it/s] 85%|████████▌ | 1064/1250 [00:19<00:01, 94.15it/s] 86%|████████▌ | 1074/1250 [00:19<00:01, 94.10it/s] 87%|████████▋ | 1084/1250 [00:19<00:01, 94.09it/s] 88%|████████▊ | 1094/1250 [00:19<00:01, 94.07it/s] 88%|████████▊ | 1104/1250 [00:19<00:01, 94.14it/s] 89%|████████▉ | 1114/1250 [00:19<00:01, 94.11it/s] 90%|████████▉ | 1124/1250 [00:19<00:01, 93.26it/s] 91%|█████████ | 1134/1250 [00:20<00:01, 93.54it/s] 92%|█████████▏| 1144/1250 [00:20<00:01, 93.75it/s] 92%|█████████▏| 1154/1250 [00:20<00:01, 93.73it/s] 93%|█████████▎| 1164/1250 [00:20<00:00, 93.84it/s] 94%|█████████▍| 1174/1250 [00:20<00:00, 93.80it/s] 95%|█████████▍| 1184/1250 [00:20<00:00, 93.50it/s] 96%|█████████▌| 1194/1250 [00:20<00:00, 93.64it/s] 96%|█████████▋| 1204/1250 [00:20<00:00, 93.79it/s] 97%|█████████▋| 1214/1250 [00:20<00:00, 93.91it/s] 98%|█████████▊| 1224/1250 [00:21<00:00, 94.01it/s] 99%|█████████▊| 1234/1250 [00:21<00:00, 94.02it/s] 100%|█████████▉| 1244/1250 [00:21<00:00, 94.09it/s] 100%|██████████| 1250/1250 [00:21<00:00, 58.57it/s]
INFO:TestAccuracy:Batch size is 4, Accuracy: 0.772
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:absl:Fingerprint not found. Saved model loading will continue.
INFO:BackendDCU:Batch size is 4, QPS: 471, Avg Latency:8.49, Tail Latency:8.81
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:absl:Fingerprint not found. Saved model loading will continue.
INFO:BackendDCU:Batch size is 8, QPS: 562, Avg Latency:14.22, Tail Latency:16.69
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:absl:Fingerprint not found. Saved model loading will continue.
INFO:BackendDCU:Batch size is 16, QPS: 768, Avg Latency:20.83, Tail Latency:21.58
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:absl:Fingerprint not found. Saved model loading will continue.
INFO:BackendDCU:Batch size is 32, QPS: 866, Avg Latency:36.94, Tail Latency:39.22
INFO:tensorflow:Saver not created because there are no variables in the graph to restore
INFO:absl:Fingerprint not found. Saved model loading will continue.
INFO:BackendDCU:Batch size is 64, QPS: 910, Avg Latency:70.26, Tail Latency:72.84
INFO:PerfEngine:Testing Finish. Report is saved in path: [ general_perf/reports/DCU/resnet50-tf-fp32/result-fp32.json ]
INFO:PerfEngine:PDF Version is saved in path: [ general_perf/reports/DCU/resnet50-tf-fp32/RESNET50-TF-FP32-TO-FP32.JSON.pdf ]
Exception ignored in: <function AtomicFunction.__del__ at 0x7f51cc184af0>
Traceback (most recent call last):
File "/usr/local/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/atomic_function.py", line 218, in __del__
TypeError: 'NoneType' object is not subscriptable
Exception ignored in: <function AtomicFunction.__del__ at 0x7f51cc184af0>
Traceback (most recent call last):
File "/usr/local/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/atomic_function.py", line 218, in __del__
TypeError: 'NoneType' object is not subscriptable
Exception ignored in: <function AtomicFunction.__del__ at 0x7f51cc184af0>
Traceback (most recent call last):
File "/usr/local/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/atomic_function.py", line 218, in __del__
TypeError: 'NoneType' object is not subscriptable
ByteMLPerf/byte_infer_perf/general_perf/log/resnet50-torch-fp16.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
/usr/local/lib/python3.10/site-packages/tensorflow/python/keras/engine/training_arrays_v1.py:37: UserWarning: A NumPy version >=1.23.5 and <2.3.0 is required for this version of SciPy (detected version 1.23.0)
from scipy.sparse import issparse # pylint: disable=g-import-not-at-top
INFO:PerfEngine:******************************************* Start to test model: resnet50-torch-fp16. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_imagenet
INFO:Imagenet:Initial...
0it [00:00, ?it/s] 8501it [00:00, 85004.61it/s] 30538it [00:00, 164622.63it/s] 50000it [00:00, 174036.35it/s]
INFO:Imagenet:reduced image list, 45000 images not found
INFO:Imagenet:loaded 5000 images, cache=0, took=0.3sec
INFO:Imagenet:Preprocessing...
INFO:Imagenet:Rebatching batch size to: 1 ...
0%| | 0/5000 [00:00<?, ?it/s] 4%|▍ | 189/5000 [00:00<00:02, 1883.84it/s] 8%|▊ | 381/5000 [00:00<00:02, 1904.60it/s] 12%|█▏ | 576/5000 [00:00<00:02, 1925.28it/s] 15%|█▌ | 773/5000 [00:00<00:02, 1940.08it/s] 19%|█▉ | 970/5000 [00:00<00:02, 1950.34it/s] 23%|██▎ | 1167/5000 [00:00<00:01, 1955.35it/s] 27%|██▋ | 1365/5000 [00:00<00:01, 1961.91it/s] 31%|███ | 1562/5000 [00:00<00:01, 1963.37it/s] 35%|███▌ | 1763/5000 [00:00<00:01, 1975.88it/s] 39%|███▉ | 1964/5000 [00:01<00:01, 1984.31it/s] 43%|████▎ | 2164/5000 [00:01<00:01, 1988.18it/s] 47%|████▋ | 2363/5000 [00:01<00:01, 1987.03it/s] 51%|█████▏ | 2563/5000 [00:01<00:01, 1989.59it/s] 55%|█████▌ | 2763/5000 [00:01<00:01, 1992.33it/s] 59%|█████▉ | 2964/5000 [00:01<00:01, 1996.99it/s] 63%|██████▎ | 3166/5000 [00:01<00:00, 2002.40it/s] 67%|██████▋ | 3367/5000 [00:01<00:00, 2002.84it/s] 71%|███████▏ | 3571/5000 [00:01<00:00, 2012.98it/s] 76%|███████▌ | 3775/5000 [00:01<00:00, 2018.87it/s] 80%|███████▉ | 3978/5000 [00:02<00:00, 2022.08it/s] 84%|████████▎ | 4182/5000 [00:02<00:00, 2026.16it/s] 88%|████████▊ | 4385/5000 [00:02<00:00, 2022.70it/s] 92%|█████████▏| 4589/5000 [00:02<00:00, 2026.83it/s] 96%|█████████▌| 4793/5000 [00:02<00:00, 2028.51it/s] 100%|█████████▉| 4997/5000 [00:02<00:00, 2030.70it/s] 100%|██████████| 5000/5000 [00:02<00:00, 1993.98it/s]
INFO:PerfEngine:Start to compile the model...
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:Imagenet:Rebatching batch size to: 1 ...
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/5000 [00:00<?, ?it/s] 0%| | 1/5000 [00:04<6:28:49, 4.67s/it] 0%| | 16/5000 [00:04<17:49, 4.66it/s] 1%| | 41/5000 [00:04<05:39, 14.62it/s] 1%|▏ | 67/5000 [00:04<02:55, 28.04it/s] 2%|▏ | 96/5000 [00:05<01:44, 47.03it/s] 2%|▎ | 125/5000 [00:05<01:09, 69.80it/s] 3%|▎ | 154/5000 [00:05<00:50, 95.77it/s] 4%|▎ | 183/5000 [00:05<00:39, 123.31it/s] 4%|▍ | 212/5000 [00:05<00:31, 151.48it/s] 5%|▍ | 241/5000 [00:05<00:26, 178.30it/s] 5%|▌ | 270/5000 [00:05<00:23, 202.26it/s] 6%|▌ | 299/5000 [00:05<00:21, 222.33it/s] 7%|▋ | 328/5000 [00:05<00:19, 238.73it/s] 7%|▋ | 357/5000 [00:05<00:18, 251.49it/s] 8%|▊ | 386/5000 [00:06<00:17, 261.21it/s] 8%|▊ | 415/5000 [00:06<00:17, 268.49it/s] 9%|▉ | 444/5000 [00:06<00:16, 273.56it/s] 9%|▉ | 473/5000 [00:06<00:16, 277.21it/s] 10%|█ | 502/5000 [00:06<00:16, 279.78it/s] 11%|█ | 531/5000 [00:06<00:15, 281.42it/s] 11%|█ | 560/5000 [00:06<00:15, 280.85it/s] 12%|█▏ | 589/5000 [00:06<00:15, 282.25it/s] 12%|█▏ | 618/5000 [00:06<00:15, 283.12it/s] 13%|█▎ | 647/5000 [00:07<00:15, 283.87it/s] 14%|█▎ | 676/5000 [00:07<00:15, 284.28it/s] 14%|█▍ | 705/5000 [00:07<00:15, 284.53it/s] 15%|█▍ | 734/5000 [00:07<00:14, 284.58it/s] 15%|█▌ | 763/5000 [00:07<00:14, 284.57it/s] 16%|█▌ | 792/5000 [00:07<00:14, 284.57it/s] 16%|█▋ | 821/5000 [00:07<00:14, 284.59it/s] 17%|█▋ | 850/5000 [00:07<00:14, 284.75it/s] 18%|█▊ | 879/5000 [00:07<00:14, 284.07it/s] 18%|█▊ | 908/5000 [00:07<00:14, 284.08it/s] 19%|█▊ | 937/5000 [00:08<00:14, 282.23it/s] 19%|█▉ | 966/5000 [00:08<00:14, 282.91it/s] 20%|█▉ | 995/5000 [00:08<00:14, 283.54it/s] 20%|██ | 1024/5000 [00:08<00:14, 283.91it/s] 21%|██ | 1053/5000 [00:08<00:13, 284.11it/s] 22%|██▏ | 1082/5000 [00:08<00:13, 284.39it/s] 22%|██▏ | 1111/5000 [00:08<00:13, 284.72it/s] 23%|██▎ | 1140/5000 [00:08<00:13, 284.76it/s] 23%|██▎ | 1169/5000 [00:08<00:13, 284.82it/s] 24%|██▍ | 1198/5000 [00:08<00:13, 284.77it/s] 25%|██▍ | 1227/5000 [00:09<00:13, 284.89it/s] 25%|██▌ | 1256/5000 [00:09<00:13, 284.97it/s] 26%|██▌ | 1285/5000 [00:09<00:13, 284.88it/s] 26%|██▋ | 1314/5000 [00:09<00:13, 282.73it/s] 27%|██▋ | 1343/5000 [00:09<00:12, 283.42it/s] 27%|██▋ | 1372/5000 [00:09<00:12, 283.87it/s] 28%|██▊ | 1401/5000 [00:09<00:12, 284.29it/s] 29%|██▊ | 1430/5000 [00:09<00:12, 284.55it/s] 29%|██▉ | 1459/5000 [00:09<00:12, 284.67it/s] 30%|██▉ | 1488/5000 [00:09<00:12, 284.90it/s] 30%|███ | 1517/5000 [00:10<00:12, 284.84it/s] 31%|███ | 1546/5000 [00:10<00:12, 284.57it/s] 32%|███▏ | 1575/5000 [00:10<00:12, 284.58it/s] 32%|███▏ | 1604/5000 [00:10<00:11, 284.75it/s] 33%|███▎ | 1633/5000 [00:10<00:11, 284.76it/s] 33%|███▎ | 1662/5000 [00:10<00:11, 284.66it/s] 34%|███▍ | 1691/5000 [00:10<00:11, 282.96it/s] 34%|███▍ | 1720/5000 [00:10<00:11, 283.49it/s] 35%|███▍ | 1749/5000 [00:10<00:11, 283.67it/s] 36%|███▌ | 1778/5000 [00:10<00:11, 284.03it/s] 36%|███▌ | 1807/5000 [00:11<00:11, 284.05it/s] 37%|███▋ | 1836/5000 [00:11<00:11, 284.03it/s] 37%|███▋ | 1865/5000 [00:11<00:11, 284.04it/s] 38%|███▊ | 1894/5000 [00:11<00:10, 283.95it/s] 38%|███▊ | 1923/5000 [00:11<00:10, 284.04it/s] 39%|███▉ | 1952/5000 [00:11<00:10, 284.27it/s] 40%|███▉ | 1981/5000 [00:11<00:10, 284.17it/s] 40%|████ | 2010/5000 [00:11<00:10, 284.19it/s] 41%|████ | 2039/5000 [00:11<00:10, 284.31it/s] 41%|████▏ | 2068/5000 [00:12<00:10, 282.28it/s] 42%|████▏ | 2097/5000 [00:12<00:10, 282.65it/s] 43%|████▎ | 2126/5000 [00:12<00:10, 282.99it/s] 43%|████▎ | 2155/5000 [00:12<00:10, 283.27it/s] 44%|████▎ | 2184/5000 [00:12<00:09, 283.55it/s] 44%|████▍ | 2213/5000 [00:12<00:09, 283.56it/s] 45%|████▍ | 2242/5000 [00:12<00:09, 283.67it/s] 45%|████▌ | 2271/5000 [00:12<00:09, 283.83it/s] 46%|████▌ | 2300/5000 [00:12<00:09, 283.87it/s] 47%|████▋ | 2329/5000 [00:12<00:09, 283.73it/s] 47%|████▋ | 2358/5000 [00:13<00:09, 283.65it/s] 48%|████▊ | 2387/5000 [00:13<00:09, 283.59it/s] 48%|████▊ | 2416/5000 [00:13<00:09, 283.62it/s] 49%|████▉ | 2445/5000 [00:13<00:09, 283.78it/s] 49%|████▉ | 2474/5000 [00:13<00:08, 281.86it/s] 50%|█████ | 2503/5000 [00:13<00:08, 281.94it/s] 51%|█████ | 2532/5000 [00:13<00:08, 282.45it/s] 51%|█████ | 2561/5000 [00:13<00:08, 282.94it/s] 52%|█████▏ | 2590/5000 [00:13<00:08, 283.37it/s] 52%|█████▏ | 2619/5000 [00:13<00:08, 283.60it/s] 53%|█████▎ | 2648/5000 [00:14<00:08, 283.45it/s] 54%|█████▎ | 2677/5000 [00:14<00:08, 283.33it/s] 54%|█████▍ | 2706/5000 [00:14<00:08, 283.46it/s] 55%|█████▍ | 2735/5000 [00:14<00:07, 283.57it/s] 55%|█████▌ | 2764/5000 [00:14<00:07, 283.65it/s] 56%|█████▌ | 2793/5000 [00:14<00:07, 283.45it/s] 56%|█████▋ | 2822/5000 [00:14<00:07, 283.64it/s] 57%|█████▋ | 2851/5000 [00:14<00:07, 281.77it/s] 58%|█████▊ | 2880/5000 [00:14<00:07, 282.28it/s] 58%|█████▊ | 2909/5000 [00:14<00:07, 281.98it/s] 59%|█████▉ | 2938/5000 [00:15<00:07, 282.55it/s] 59%|█████▉ | 2967/5000 [00:15<00:07, 282.71it/s] 60%|█████▉ | 2996/5000 [00:15<00:07, 282.82it/s] 60%|██████ | 3025/5000 [00:15<00:06, 282.60it/s] 61%|██████ | 3054/5000 [00:15<00:06, 282.77it/s] 62%|██████▏ | 3083/5000 [00:15<00:06, 283.31it/s] 62%|██████▏ | 3112/5000 [00:15<00:06, 283.57it/s] 63%|██████▎ | 3141/5000 [00:15<00:06, 283.65it/s] 63%|██████▎ | 3170/5000 [00:15<00:06, 283.68it/s] 64%|██████▍ | 3199/5000 [00:15<00:06, 283.61it/s] 65%|██████▍ | 3228/5000 [00:16<00:06, 281.45it/s] 65%|██████▌ | 3257/5000 [00:16<00:06, 281.95it/s] 66%|██████▌ | 3286/5000 [00:16<00:06, 282.27it/s] 66%|██████▋ | 3315/5000 [00:16<00:05, 282.52it/s] 67%|██████▋ | 3344/5000 [00:16<00:05, 282.61it/s] 67%|██████▋ | 3373/5000 [00:16<00:05, 282.75it/s] 68%|██████▊ | 3402/5000 [00:16<00:05, 282.89it/s] 69%|██████▊ | 3431/5000 [00:16<00:05, 282.88it/s] 69%|██████▉ | 3460/5000 [00:16<00:05, 282.68it/s] 70%|██████▉ | 3489/5000 [00:17<00:05, 282.59it/s] 70%|███████ | 3518/5000 [00:17<00:05, 282.39it/s] 71%|███████ | 3547/5000 [00:17<00:05, 282.49it/s] 72%|███████▏ | 3576/5000 [00:17<00:05, 282.58it/s] 72%|███████▏ | 3605/5000 [00:17<00:04, 280.20it/s] 73%|███████▎ | 3634/5000 [00:17<00:04, 280.93it/s] 73%|███████▎ | 3663/5000 [00:17<00:04, 281.33it/s] 74%|███████▍ | 3692/5000 [00:17<00:04, 281.59it/s] 74%|███████▍ | 3721/5000 [00:17<00:04, 281.77it/s] 75%|███████▌ | 3750/5000 [00:17<00:04, 281.84it/s] 76%|███████▌ | 3779/5000 [00:18<00:04, 281.97it/s] 76%|███████▌ | 3808/5000 [00:18<00:04, 281.86it/s] 77%|███████▋ | 3837/5000 [00:18<00:04, 282.04it/s] 77%|███████▋ | 3866/5000 [00:18<00:04, 281.92it/s] 78%|███████▊ | 3895/5000 [00:18<00:03, 281.88it/s] 78%|███████▊ | 3924/5000 [00:18<00:03, 281.78it/s] 79%|███████▉ | 3953/5000 [00:18<00:03, 281.94it/s] 80%|███████▉ | 3982/5000 [00:18<00:03, 279.83it/s] 80%|████████ | 4011/5000 [00:18<00:03, 280.17it/s] 81%|████████ | 4040/5000 [00:18<00:03, 280.45it/s] 81%|████████▏ | 4069/5000 [00:19<00:03, 280.70it/s] 82%|████████▏ | 4098/5000 [00:19<00:03, 280.79it/s] 83%|████████▎ | 4127/5000 [00:19<00:03, 280.84it/s] 83%|████████▎ | 4156/5000 [00:19<00:03, 280.99it/s] 84%|████████▎ | 4185/5000 [00:19<00:02, 280.90it/s] 84%|████████▍ | 4214/5000 [00:19<00:02, 280.85it/s] 85%|████████▍ | 4243/5000 [00:19<00:02, 281.09it/s] 85%|████████▌ | 4272/5000 [00:19<00:02, 281.22it/s] 86%|████████▌ | 4301/5000 [00:19<00:02, 281.22it/s] 87%|████████▋ | 4330/5000 [00:20<00:02, 281.32it/s] 87%|████████▋ | 4359/5000 [00:20<00:02, 279.47it/s] 88%|████████▊ | 4388/5000 [00:20<00:02, 279.93it/s] 88%|████████▊ | 4417/5000 [00:20<00:02, 280.14it/s] 89%|████████▉ | 4446/5000 [00:20<00:01, 280.44it/s] 90%|████████▉ | 4475/5000 [00:20<00:01, 280.61it/s] 90%|█████████ | 4504/5000 [00:20<00:01, 280.47it/s] 91%|█████████ | 4533/5000 [00:20<00:01, 280.66it/s] 91%|█████████ | 4562/5000 [00:20<00:01, 280.59it/s] 92%|█████████▏| 4591/5000 [00:20<00:01, 280.68it/s] 92%|█████████▏| 4620/5000 [00:21<00:01, 280.46it/s] 93%|█████████▎| 4649/5000 [00:21<00:01, 280.68it/s] 94%|█████████▎| 4678/5000 [00:21<00:01, 280.51it/s] 94%|█████████▍| 4707/5000 [00:21<00:01, 280.41it/s] 95%|█████████▍| 4736/5000 [00:21<00:00, 278.89it/s] 95%|█████████▌| 4764/5000 [00:21<00:00, 279.19it/s] 96%|█████████▌| 4793/5000 [00:21<00:00, 279.52it/s] 96%|█████████▋| 4821/5000 [00:21<00:00, 279.65it/s] 97%|█████████▋| 4849/5000 [00:21<00:00, 279.54it/s] 98%|█████████▊| 4878/5000 [00:21<00:00, 279.78it/s] 98%|█████████▊| 4907/5000 [00:22<00:00, 280.04it/s] 99%|█████████▊| 4936/5000 [00:22<00:00, 279.90it/s] 99%|█████████▉| 4965/5000 [00:22<00:00, 279.93it/s] 100%|█████████▉| 4993/5000 [00:22<00:00, 279.82it/s] 100%|██████████| 5000/5000 [00:22<00:00, 223.13it/s]
INFO:TestAccuracy:Batch size is 1, Accuracy: 0.7698
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
INFO:BackendDCU:Batch size is 1, QPS: 301, Avg Latency:3.32, Tail Latency:3.36
INFO:BackendDCU:Batch size is 2, QPS: 536, Avg Latency:3.73, Tail Latency:3.78
INFO:BackendDCU:Batch size is 4, QPS: 852, Avg Latency:4.69, Tail Latency:4.91
INFO:BackendDCU:Batch size is 8, QPS: 1169, Avg Latency:6.84, Tail Latency:9.08
INFO:BackendDCU:Batch size is 16, QPS: 1542, Avg Latency:10.37, Tail Latency:12.63
INFO:BackendDCU:Batch size is 32, QPS: 1800, Avg Latency:17.77, Tail Latency:20.28
INFO:BackendDCU:Batch size is 64, QPS: 1946, Avg Latency:32.88, Tail Latency:37.13
INFO:BackendDCU:Batch size is 128, QPS: 1736, Avg Latency:73.72, Tail Latency:76.02
INFO:BackendDCU:Batch size is 256, QPS: 1782, Avg Latency:143.58, Tail Latency:146.29
INFO:BackendDCU:Batch size is 512, QPS: 1972, Avg Latency:259.56, Tail Latency:262.09
INFO:BackendDCU:Batch size is 1024, QPS: 2065, Avg Latency:495.8, Tail Latency:499.68
INFO:PerfEngine:Testing Finish. Report is saved in path: [ general_perf/reports/DCU/resnet50-torch-fp16/result-fp16.json ]
INFO:PerfEngine:PDF Version is saved in path: [ general_perf/reports/DCU/resnet50-torch-fp16/RESNET50-TORCH-FP16-TO-FP16.JSON.pdf ]
ByteMLPerf/byte_infer_perf/general_perf/log/resnet50-torch-fp32.log 0 → 100644
View file @ 24b257f1
INFO:LANUCH:******************* Pip Package Installing *******************
INFO:PerfEngine:******************* Backend Env Initization *******************
INFO:BackendStore:Loading Compile Backend: DCU
INFO:BackendStore:Loading Runtime Backend: DCU
INFO:PerfEngine:******************************************* Start to test model: resnet50-torch-fp32. *******************************************
INFO:PerfEngine:******************************************* Running Backend Compilation... *******************************************
INFO:PerfEngine:Running Backend Preoptimization...
INFO:DatasetStore:Loading Dataset: open_imagenet
INFO:Imagenet:Initial...
0it [00:00, ?it/s] 8901it [00:00, 89004.99it/s] 31260it [00:00, 168165.72it/s] 50000it [00:00, 176767.95it/s]
INFO:Imagenet:reduced image list, 45000 images not found
INFO:Imagenet:loaded 5000 images, cache=0, took=0.3sec
INFO:Imagenet:Preprocessing...
INFO:Imagenet:Rebatching batch size to: 1 ...
0%| | 0/5000 [00:00<?, ?it/s] 4%|▎ | 186/5000 [00:00<00:02, 1858.65it/s] 8%|▊ | 375/5000 [00:00<00:02, 1875.83it/s] 11%|█▏ | 563/5000 [00:00<00:02, 1869.25it/s] 15%|█▌ | 750/5000 [00:00<00:02, 1866.08it/s] 19%|█▊ | 937/5000 [00:00<00:02, 1863.84it/s] 22%|██▎ | 1125/5000 [00:00<00:02, 1869.22it/s] 26%|██▋ | 1314/5000 [00:00<00:01, 1873.91it/s] 30%|███ | 1504/5000 [00:00<00:01, 1879.82it/s] 34%|███▍ | 1696/5000 [00:00<00:01, 1889.78it/s] 38%|███▊ | 1888/5000 [00:01<00:01, 1898.17it/s] 42%|████▏ | 2078/5000 [00:01<00:01, 1898.25it/s] 45%|████▌ | 2271/5000 [00:01<00:01, 1906.96it/s] 49%|████▉ | 2464/5000 [00:01<00:01, 1912.94it/s] 53%|█████▎ | 2656/5000 [00:01<00:01, 1902.19it/s] 57%|█████▋ | 2847/5000 [00:01<00:01, 1901.69it/s] 61%|██████ | 3039/5000 [00:01<00:01, 1905.57it/s] 65%|██████▍ | 3231/5000 [00:01<00:00, 1909.49it/s] 68%|██████▊ | 3423/5000 [00:01<00:00, 1911.32it/s] 72%|███████▏ | 3615/5000 [00:01<00:00, 1910.36it/s] 76%|███████▌ | 3809/5000 [00:02<00:00, 1917.51it/s] 80%|████████ | 4003/5000 [00:02<00:00, 1922.62it/s] 84%|████████▍ | 4198/5000 [00:02<00:00, 1929.01it/s] 88%|████████▊ | 4394/5000 [00:02<00:00, 1936.10it/s] 92%|█████████▏| 4591/5000 [00:02<00:00, 1945.75it/s] 96%|█████████▌| 4786/5000 [00:02<00:00, 1944.98it/s] 100%|█████████▉| 4981/5000 [00:02<00:00, 1944.73it/s] 100%|██████████| 5000/5000 [00:02<00:00, 1908.29it/s]
INFO:PerfEngine:Start to compile the model...
INFO:PerfEngine:******************************************* Running Accuracy Checker... *******************************************
INFO:Imagenet:Rebatching batch size to: 1 ...
INFO:TestAccuracy:Start to calculate accuracy...
0%| | 0/5000 [00:00<?, ?it/s] 0%| | 1/5000 [00:05<8:04:12, 5.81s/it] 0%| | 19/5000 [00:05<18:30, 4.48it/s] 1%| | 48/5000 [00:06<05:55, 13.93it/s] 2%|▏ | 77/5000 [00:06<03:06, 26.36it/s] 2%|▏ | 106/5000 [00:06<01:55, 42.25it/s] 3%|▎ | 135/5000 [00:06<01:18, 61.81it/s] 3%|▎ | 164/5000 [00:06<00:57, 84.82it/s] 4%|▍ | 193/5000 [00:06<00:43, 110.42it/s] 4%|▍ | 222/5000 [00:06<00:34, 137.26it/s] 5%|▌ | 251/5000 [00:06<00:29, 163.75it/s] 6%|▌ | 280/5000 [00:06<00:25, 188.28it/s] 6%|▌ | 309/5000 [00:06<00:22, 209.77it/s] 7%|▋ | 337/5000 [00:07<00:20, 226.02it/s] 7%|▋ | 366/5000 [00:07<00:19, 240.64it/s] 8%|▊ | 395/5000 [00:07<00:18, 252.09it/s] 8%|▊ | 424/5000 [00:07<00:17, 260.68it/s] 9%|▉ | 453/5000 [00:07<00:17, 266.95it/s] 10%|▉ | 482/5000 [00:07<00:16, 271.53it/s] 10%|█ | 511/5000 [00:07<00:16, 274.77it/s] 11%|█ | 540/5000 [00:07<00:16, 277.17it/s] 11%|█▏ | 569/5000 [00:07<00:15, 278.70it/s] 12%|█▏ | 598/5000 [00:07<00:15, 279.80it/s] 13%|█▎ | 627/5000 [00:08<00:15, 280.57it/s] 13%|█▎ | 656/5000 [00:08<00:15, 281.12it/s] 14%|█▎ | 685/5000 [00:08<00:15, 281.38it/s] 14%|█▍ | 714/5000 [00:08<00:15, 279.78it/s] 15%|█▍ | 743/5000 [00:08<00:15, 280.47it/s] 15%|█▌ | 772/5000 [00:08<00:15, 281.18it/s] 16%|█▌ | 801/5000 [00:08<00:14, 281.44it/s] 17%|█▋ | 830/5000 [00:08<00:14, 281.72it/s] 17%|█▋ | 859/5000 [00:08<00:14, 281.74it/s] 18%|█▊ | 888/5000 [00:08<00:14, 281.83it/s] 18%|█▊ | 917/5000 [00:09<00:14, 281.88it/s] 19%|█▉ | 946/5000 [00:09<00:14, 282.03it/s] 20%|█▉ | 975/5000 [00:09<00:14, 282.07it/s] 20%|██ | 1004/5000 [00:09<00:14, 282.21it/s] 21%|██ | 1033/5000 [00:09<00:14, 282.13it/s] 21%|██ | 1062/5000 [00:09<00:13, 282.32it/s] 22%|██▏ | 1091/5000 [00:09<00:13, 280.67it/s] 22%|██▏ | 1120/5000 [00:09<00:13, 281.28it/s] 23%|██▎ | 1149/5000 [00:09<00:13, 281.74it/s] 24%|██▎ | 1178/5000 [00:10<00:13, 282.08it/s] 24%|██▍ | 1207/5000 [00:10<00:13, 282.25it/s] 25%|██▍ | 1236/5000 [00:10<00:13, 282.38it/s] 25%|██▌ | 1265/5000 [00:10<00:13, 282.52it/s] 26%|██▌ | 1294/5000 [00:10<00:13, 282.62it/s] 26%|██▋ | 1323/5000 [00:10<00:13, 282.62it/s] 27%|██▋ | 1352/5000 [00:10<00:12, 282.56it/s] 28%|██▊ | 1381/5000 [00:10<00:12, 282.55it/s] 28%|██▊ | 1410/5000 [00:10<00:12, 282.49it/s] 29%|██▉ | 1439/5000 [00:10<00:12, 282.53it/s] 29%|██▉ | 1468/5000 [00:11<00:12, 280.64it/s] 30%|██▉ | 1497/5000 [00:11<00:12, 281.30it/s] 31%|███ | 1526/5000 [00:11<00:12, 281.66it/s] 31%|███ | 1555/5000 [00:11<00:12, 281.99it/s] 32%|███▏ | 1584/5000 [00:11<00:12, 282.24it/s] 32%|███▏ | 1613/5000 [00:11<00:11, 282.63it/s] 33%|███▎ | 1642/5000 [00:11<00:11, 282.68it/s] 33%|███▎ | 1671/5000 [00:11<00:11, 282.61it/s] 34%|███▍ | 1700/5000 [00:11<00:11, 282.77it/s] 35%|███▍ | 1729/5000 [00:11<00:11, 282.73it/s] 35%|███▌ | 1758/5000 [00:12<00:11, 282.75it/s] 36%|███▌ | 1787/5000 [00:12<00:11, 282.75it/s] 36%|███▋ | 1816/5000 [00:12<00:11, 282.77it/s] 37%|███▋ | 1845/5000 [00:12<00:11, 280.72it/s] 37%|███▋ | 1874/5000 [00:12<00:11, 281.37it/s] 38%|███▊ | 1903/5000 [00:12<00:10, 281.85it/s] 39%|███▊ | 1932/5000 [00:12<00:10, 281.98it/s] 39%|███▉ | 1961/5000 [00:12<00:10, 282.19it/s] 40%|███▉ | 1990/5000 [00:12<00:10, 282.28it/s] 40%|████ | 2019/5000 [00:13<00:10, 282.30it/s] 41%|████ | 2048/5000 [00:13<00:10, 282.45it/s] 42%|████▏ | 2077/5000 [00:13<00:10, 282.40it/s] 42%|████▏ | 2106/5000 [00:13<00:10, 282.48it/s] 43%|████▎ | 2135/5000 [00:13<00:10, 282.54it/s] 43%|████▎ | 2164/5000 [00:13<00:10, 282.42it/s] 44%|████▍ | 2193/5000 [00:13<00:09, 282.43it/s] 44%|████▍ | 2222/5000 [00:13<00:09, 280.41it/s] 45%|████▌ | 2251/5000 [00:13<00:09, 281.06it/s] 46%|████▌ | 2280/5000 [00:13<00:09, 281.45it/s] 46%|████▌ | 2309/5000 [00:14<00:09, 281.83it/s] 47%|████▋ | 2338/5000 [00:14<00:09, 282.15it/s] 47%|████▋ | 2367/5000 [00:14<00:09, 282.14it/s] 48%|████▊ | 2396/5000 [00:14<00:09, 282.24it/s] 48%|████▊ | 2425/5000 [00:14<00:09, 282.23it/s] 49%|████▉ | 2454/5000 [00:14<00:09, 282.23it/s] 50%|████▉ | 2483/5000 [00:14<00:08, 282.31it/s] 50%|█████ | 2512/5000 [00:14<00:08, 282.42it/s] 51%|█████ | 2541/5000 [00:14<00:08, 282.34it/s] 51%|█████▏ | 2570/5000 [00:14<00:08, 282.27it/s] 52%|█████▏ | 2599/5000 [00:15<00:08, 282.18it/s] 53%|█████▎ | 2628/5000 [00:15<00:08, 280.40it/s] 53%|█████▎ | 2657/5000 [00:15<00:08, 281.05it/s] 54%|█████▎ | 2686/5000 [00:15<00:08, 281.56it/s] 54%|█████▍ | 2715/5000 [00:15<00:08, 281.67it/s] 55%|█████▍ | 2744/5000 [00:15<00:08, 281.97it/s] 55%|█████▌ | 2773/5000 [00:15<00:07, 282.01it/s] 56%|█████▌ | 2802/5000 [00:15<00:07, 282.11it/s] 57%|█████▋ | 2831/5000 [00:15<00:07, 282.13it/s] 57%|█████▋ | 2860/5000 [00:15<00:07, 282.09it/s] 58%|█████▊ | 2889/5000 [00:16<00:07, 281.95it/s] 58%|█████▊ | 2918/5000 [00:16<00:07, 281.80it/s] 59%|█████▉ | 2947/5000 [00:16<00:07, 281.57it/s] 60%|█████▉ | 2976/5000 [00:16<00:07, 281.44it/s] 60%|██████ | 3005/5000 [00:16<00:07, 279.74it/s] 61%|██████ | 3034/5000 [00:16<00:07, 280.30it/s] 61%|██████▏ | 3063/5000 [00:16<00:06, 280.52it/s] 62%|██████▏ | 3092/5000 [00:16<00:06, 280.59it/s] 62%|██████▏ | 3121/5000 [00:16<00:06, 280.73it/s] 63%|██████▎ | 3150/5000 [00:17<00:06, 280.92it/s] 64%|██████▎ | 3179/5000 [00:17<00:06, 279.81it/s] 64%|██████▍ | 3208/5000 [00:17<00:06, 280.24it/s] 65%|██████▍ | 3237/5000 [00:17<00:06, 280.65it/s] 65%|██████▌ | 3266/5000 [00:17<00:06, 280.90it/s] 66%|██████▌ | 3295/5000 [00:17<00:06, 281.18it/s] 66%|██████▋ | 3324/5000 [00:17<00:05, 281.35it/s] 67%|██████▋ | 3353/5000 [00:17<00:05, 281.51it/s] 68%|██████▊ | 3382/5000 [00:17<00:05, 279.61it/s] 68%|██████▊ | 3411/5000 [00:17<00:05, 280.51it/s] 69%|██████▉ | 3440/5000 [00:18<00:05, 280.98it/s] 69%|██████▉ | 3469/5000 [00:18<00:05, 281.08it/s] 70%|██████▉ | 3498/5000 [00:18<00:05, 281.04it/s] 71%|███████ | 3527/5000 [00:18<00:05, 281.25it/s] 71%|███████ | 3556/5000 [00:18<00:05, 281.42it/s] 72%|███████▏ | 3585/5000 [00:18<00:05, 281.63it/s] 72%|███████▏ | 3614/5000 [00:18<00:04, 281.64it/s] 73%|███████▎ | 3643/5000 [00:18<00:04, 281.72it/s] 73%|███████▎ | 3672/5000 [00:18<00:04, 281.64it/s] 74%|███████▍ | 3701/5000 [00:18<00:04, 281.69it/s] 75%|███████▍ | 3730/5000 [00:19<00:04, 281.63it/s] 75%|███████▌ | 3759/5000 [00:19<00:04, 279.72it/s] 76%|███████▌ | 3788/5000 [00:19<00:04, 280.30it/s] 76%|███████▋ | 3817/5000 [00:19<00:04, 280.62it/s] 77%|███████▋ | 3846/5000 [00:19<00:04, 280.86it/s] 78%|███████▊ | 3875/5000 [00:19<00:04, 281.08it/s] 78%|███████▊ | 3904/5000 [00:19<00:03, 281.32it/s] 79%|███████▊ | 3933/5000 [00:19<00:03, 281.38it/s] 79%|███████▉ | 3962/5000 [00:19<00:03, 281.44it/s] 80%|███████▉ | 3991/5000 [00:20<00:03, 281.65it/s] 80%|████████ | 4020/5000 [00:20<00:03, 281.61it/s] 81%|████████ | 4049/5000 [00:20<00:03, 281.47it/s] 82%|████████▏ | 4078/5000 [00:20<00:03, 281.43it/s] 82%|████████▏ | 4107/5000 [00:20<00:03, 281.43it/s] 83%|████████▎ | 4136/5000 [00:20<00:03, 279.55it/s] 83%|████████▎ | 4165/5000 [00:20<00:02, 279.86it/s] 84%|████████▍ | 4194/5000 [00:20<00:02, 280.46it/s] 84%|████████▍ | 4223/5000 [00:20<00:02, 280.60it/s] 85%|████████▌ | 4252/5000 [00:20<00:02, 280.85it/s] 86%|████████▌ | 4281/5000 [00:21<00:02, 280.99it/s] 86%|████████▌ | 4310/5000 [00:21<00:02, 281.11it/s] 87%|████████▋ | 4339/5000 [00:21<00:02, 281.25it/s] 87%|████████▋ | 4368/5000 [00:21<00:02, 281.23it/s] 88%|████████▊ | 4397/5000 [00:21<00:02, 281.26it/s] 89%|████████▊ | 4426/5000 [00:21<00:02, 281.30it/s] 89%|████████▉ | 4455/5000 [00:21<00:01, 281.31it/s] 90%|████████▉ | 4484/5000 [00:21<00:01, 281.19it/s] 90%|█████████ | 4513/5000 [00:21<00:01, 279.40it/s] 91%|█████████ | 4542/5000 [00:21<00:01, 279.85it/s] 91%|█████████▏| 4571/5000 [00:22<00:01, 280.09it/s] 92%|█████████▏| 4600/5000 [00:22<00:01, 280.10it/s] 93%|█████████▎| 4629/5000 [00:22<00:01, 280.42it/s] 93%|█████████▎| 4658/5000 [00:22<00:01, 280.38it/s] 94%|█████████▎| 4687/5000 [00:22<00:01, 280.50it/s] 94%|█████████▍| 4716/5000 [00:22<00:01, 280.63it/s] 95%|█████████▍| 4745/5000 [00:22<00:00, 280.70it/s] 95%|█████████▌| 4774/5000 [00:22<00:00, 280.78it/s] 96%|█████████▌| 4803/5000 [00:22<00:00, 281.06it/s] 97%|█████████▋| 4832/5000 [00:23<00:00, 280.91it/s] 97%|█████████▋| 4861/5000 [00:23<00:00, 279.15it/s] 98%|█████████▊| 4890/5000 [00:23<00:00, 279.51it/s] 98%|█████████▊| 4919/5000 [00:23<00:00, 279.68it/s] 99%|█████████▉| 4948/5000 [00:23<00:00, 280.01it/s] 100%|█████████▉| 4977/5000 [00:23<00:00, 280.05it/s] 100%|██████████| 5000/5000 [00:23<00:00, 211.78it/s]
INFO:TestAccuracy:Batch size is 1, Accuracy: 0.7696
INFO:PerfEngine:******************************************* Runing QPS Checker... *******************************************
INFO:BackendDCU:Batch size is 1, QPS: 291, Avg Latency:3.43, Tail Latency:3.47
INFO:BackendDCU:Batch size is 2, QPS: 467, Avg Latency:4.28, Tail Latency:4.47
INFO:BackendDCU:Batch size is 4, QPS: 679, Avg Latency:5.89, Tail Latency:8.23
INFO:BackendDCU:Batch size is 8, QPS: 806, Avg Latency:9.92, Tail Latency:12.15
INFO:BackendDCU:Batch size is 16, QPS: 945, Avg Latency:16.93, Tail Latency:19.09
INFO:BackendDCU:Batch size is 32, QPS: 1107, Avg Latency:28.9, Tail Latency:31.13
INFO:BackendDCU:Batch size is 64, QPS: 978, Avg Latency:65.41, Tail Latency:67.24
INFO:BackendDCU:Batch size is 128, QPS: 990, Avg Latency:129.2, Tail Latency:131.9
INFO:BackendDCU:Batch size is 256, QPS: 1036, Avg Latency:247.07, Tail Latency:249.65
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment