Skip to content

P
PaddleOCR

weixin_41840029 / PaddleOCR
与 Fork 源项目一致

Fork自 PaddlePaddle / PaddleOCR

通知 1
Star 1
Fork 0
P
PaddleOCR
未验证 提交 9c0a4d9d 编写于 作者: D Double_V 提交者： GitHub
浏览文件
操作

Merge pull request #5536 from LDOUBLEV/dygraph 

[benchmark] fix pretrain model download
上级 89480580 253c7d82
显示空白变更内容
内联 并排
Showing with 132 addition and 49 deletion
benchmark/analysis.py
浏览文件 @ 9c0a4d9d
...@@ -26,35 +26,57 @@ def parse_args(): ...@@ -26,35 +26,57 @@ def parse_args():
parser.add_argument( parser.add_argument(
"--filename", type=str, help="The name of log which need to analysis.") "--filename", type=str, help="The name of log which need to analysis.")
parser.add_argument( parser.add_argument(
"--log_with_profiler", type=str, help="The path of train log with profiler") "--log_with_profiler",
type=str,
help="The path of train log with profiler")
parser.add_argument( parser.add_argument(
"--profiler_path", type=str, help="The path of profiler timeline log.") "--profiler_path", type=str, help="The path of profiler timeline log.")
parser.add_argument( parser.add_argument(
"--keyword", type=str, help="Keyword to specify analysis data") "--keyword", type=str, help="Keyword to specify analysis data")
parser.add_argument( parser.add_argument(
"--separator", type=str, default=None, help="Separator of different field in log") "--separator",
type=str,
default=None,
help="Separator of different field in log")
parser.add_argument( parser.add_argument(
'--position', type=int, default=None, help='The position of data field') '--position', type=int, default=None, help='The position of data field')
parser.add_argument( parser.add_argument(
'--range', type=str, default="", help='The range of data field to intercept') '--range',
type=str,
default="",
help='The range of data field to intercept')
parser.add_argument( parser.add_argument(
'--base_batch_size', type=int, help='base_batch size on gpu') '--base_batch_size', type=int, help='base_batch size on gpu')
parser.add_argument( parser.add_argument(
'--skip_steps', type=int, default=0, help='The number of steps to be skipped') '--skip_steps',
type=int,
default=0,
help='The number of steps to be skipped')
parser.add_argument( parser.add_argument(
'--model_mode', type=int, default=-1, help='Analysis mode, default value is -1') '--model_mode',
type=int,
default=-1,
help='Analysis mode, default value is -1')
parser.add_argument('--ips_unit', type=str, default=None, help='IPS unit')
parser.add_argument( parser.add_argument(
'--ips_unit', type=str, default=None, help='IPS unit') '--model_name',
parser.add_argument( type=str,
'--model_name', type=str, default=0, help='training model_name, transformer_base') default=0,
help='training model_name, transformer_base')
parser.add_argument( parser.add_argument(
'--mission_name', type=str, default=0, help='training mission name') '--mission_name', type=str, default=0, help='training mission name')
parser.add_argument( parser.add_argument(
'--direction_id', type=int, default=0, help='training direction_id') '--direction_id', type=int, default=0, help='training direction_id')
parser.add_argument( parser.add_argument(
'--run_mode', type=str, default="sp", help='multi process or single process') '--run_mode',
type=str,
default="sp",
help='multi process or single process')
parser.add_argument( parser.add_argument(
'--index', type=int, default=1, help='{1: speed, 2:mem, 3:profiler, 6:max_batch_size}') '--index',
type=int,
default=1,
help='{1: speed, 2:mem, 3:profiler, 6:max_batch_size}')
parser.add_argument( parser.add_argument(
'--gpu_num', type=int, default=1, help='nums of training gpus') '--gpu_num', type=int, default=1, help='nums of training gpus')
args = parser.parse_args() args = parser.parse_args()
...@@ -72,7 +94,12 @@ def _is_number(num): ...@@ -72,7 +94,12 @@ def _is_number(num):
class TimeAnalyzer(object): class TimeAnalyzer(object):
def __init__(self, filename, keyword=None, separator=None, position=None, range="-1"): def __init__(self,
filename,
keyword=None,
separator=None,
position=None,
range="-1"):
if filename is None: if filename is None:
raise Exception("Please specify the filename!") raise Exception("Please specify the filename!")
...@@ -99,7 +126,8 @@ class TimeAnalyzer(object): ...@@ -99,7 +126,8 @@ class TimeAnalyzer(object):
# Distil the string from a line. # Distil the string from a line.
line = line.strip() line = line.strip()
line_words = line.split(self.separator) if self.separator else line.split() line_words = line.split(
self.separator) if self.separator else line.split()
if args.position: if args.position:
result = line_words[self.position] result = line_words[self.position]
else: else:
...@@ -113,16 +141,25 @@ class TimeAnalyzer(object): ...@@ -113,16 +141,25 @@ class TimeAnalyzer(object):
if not self.range: if not self.range:
result = result[0:] result = result[0:]
elif _is_number(self.range): elif _is_number(self.range):
result = result[0: int(self.range)] result = result[0:int(self.range)]
else: else:
result = result[int(self.range.split(":")[0]): int(self.range.split(":")[1])] result = result[int(self.range.split(":")[0]):int(
self.range.split(":")[1])]
self.records.append(float(result)) self.records.append(float(result))
except Exception as exc: except Exception as exc:
print("line is: {}; separator={}; position={}".format(line, self.separator, self.position)) print("line is: {}; separator={}; position={}".format(
line, self.separator, self.position))
print("Extract {} records: separator={}; position={}".format(len(self.records), self.separator, self.position)) print("Extract {} records: separator={}; position={}".format(
len(self.records), self.separator, self.position))
def _get_fps(self, mode, batch_size, gpu_num, avg_of_records, run_mode, unit=None): def _get_fps(self,
mode,
batch_size,
gpu_num,
avg_of_records,
run_mode,
unit=None):
if mode == -1 and run_mode == 'sp': if mode == -1 and run_mode == 'sp':
assert unit, "Please set the unit when mode is -1." assert unit, "Please set the unit when mode is -1."
fps = gpu_num * avg_of_records fps = gpu_num * avg_of_records
...@@ -155,12 +192,20 @@ class TimeAnalyzer(object): ...@@ -155,12 +192,20 @@ class TimeAnalyzer(object):
return fps, unit return fps, unit
def analysis(self, batch_size, gpu_num=1, skip_steps=0, mode=-1, run_mode='sp', unit=None): def analysis(self,
batch_size,
gpu_num=1,
skip_steps=0,
mode=-1,
run_mode='sp',
unit=None):
if batch_size <= 0: if batch_size <= 0:
print("base_batch_size should larger than 0.") print("base_batch_size should larger than 0.")
return 0, '' return 0, ''
if len(self.records) <= skip_steps: # to address the condition which item of log equals to skip_steps if len(
self.records
) <= skip_steps: # to address the condition which item of log equals to skip_steps
print("no records") print("no records")
return 0, '' return 0, ''
...@@ -180,16 +225,20 @@ class TimeAnalyzer(object): ...@@ -180,16 +225,20 @@ class TimeAnalyzer(object):
skip_max = self.records[i] skip_max = self.records[i]
avg_of_records = sum_of_records / float(count) avg_of_records = sum_of_records / float(count)
avg_of_records_skipped = sum_of_records_skipped / float(count - skip_steps) avg_of_records_skipped = sum_of_records_skipped / float(count -
skip_steps)
fps, fps_unit = self._get_fps(mode, batch_size, gpu_num, avg_of_records, run_mode, unit) fps, fps_unit = self._get_fps(mode, batch_size, gpu_num, avg_of_records,
fps_skipped, _ = self._get_fps(mode, batch_size, gpu_num, avg_of_records_skipped, run_mode, unit) run_mode, unit)
fps_skipped, _ = self._get_fps(mode, batch_size, gpu_num,
avg_of_records_skipped, run_mode, unit)
if mode == -1: if mode == -1:
print("average ips of %d steps, skip 0 step:" % count) print("average ips of %d steps, skip 0 step:" % count)
print("\tAvg: %.3f %s" % (avg_of_records, fps_unit)) print("\tAvg: %.3f %s" % (avg_of_records, fps_unit))
print("\tFPS: %.3f %s" % (fps, fps_unit)) print("\tFPS: %.3f %s" % (fps, fps_unit))
if skip_steps > 0: if skip_steps > 0:
print("average ips of %d steps, skip %d steps:" % (count, skip_steps)) print("average ips of %d steps, skip %d steps:" %
(count, skip_steps))
print("\tAvg: %.3f %s" % (avg_of_records_skipped, fps_unit)) print("\tAvg: %.3f %s" % (avg_of_records_skipped, fps_unit))
print("\tMin: %.3f %s" % (skip_min, fps_unit)) print("\tMin: %.3f %s" % (skip_min, fps_unit))
print("\tMax: %.3f %s" % (skip_max, fps_unit)) print("\tMax: %.3f %s" % (skip_max, fps_unit))
...@@ -199,7 +248,8 @@ class TimeAnalyzer(object): ...@@ -199,7 +248,8 @@ class TimeAnalyzer(object):
print("\tAvg: %.3f steps/s" % avg_of_records) print("\tAvg: %.3f steps/s" % avg_of_records)
print("\tFPS: %.3f %s" % (fps, fps_unit)) print("\tFPS: %.3f %s" % (fps, fps_unit))
if skip_steps > 0: if skip_steps > 0:
print("average latency of %d steps, skip %d steps:" % (count, skip_steps)) print("average latency of %d steps, skip %d steps:" %
(count, skip_steps))
print("\tAvg: %.3f steps/s" % avg_of_records_skipped) print("\tAvg: %.3f steps/s" % avg_of_records_skipped)
print("\tMin: %.3f steps/s" % skip_min) print("\tMin: %.3f steps/s" % skip_min)
print("\tMax: %.3f steps/s" % skip_max) print("\tMax: %.3f steps/s" % skip_max)
...@@ -209,7 +259,8 @@ class TimeAnalyzer(object): ...@@ -209,7 +259,8 @@ class TimeAnalyzer(object):
print("\tAvg: %.3f s/step" % avg_of_records) print("\tAvg: %.3f s/step" % avg_of_records)
print("\tFPS: %.3f %s" % (fps, fps_unit)) print("\tFPS: %.3f %s" % (fps, fps_unit))
if skip_steps > 0: if skip_steps > 0:
print("average latency of %d steps, skip %d steps:" % (count, skip_steps)) print("average latency of %d steps, skip %d steps:" %
(count, skip_steps))
print("\tAvg: %.3f s/step" % avg_of_records_skipped) print("\tAvg: %.3f s/step" % avg_of_records_skipped)
print("\tMin: %.3f s/step" % skip_min) print("\tMin: %.3f s/step" % skip_min)
print("\tMax: %.3f s/step" % skip_max) print("\tMax: %.3f s/step" % skip_max)
...@@ -236,7 +287,8 @@ if __name__ == "__main__": ...@@ -236,7 +287,8 @@ if __name__ == "__main__":
if args.gpu_num == 1: if args.gpu_num == 1:
run_info["log_with_profiler"] = args.log_with_profiler run_info["log_with_profiler"] = args.log_with_profiler
run_info["profiler_path"] = args.profiler_path run_info["profiler_path"] = args.profiler_path
analyzer = TimeAnalyzer(args.filename, args.keyword, args.separator, args.position, args.range) analyzer = TimeAnalyzer(args.filename, args.keyword, args.separator,
args.position, args.range)
run_info["FINAL_RESULT"], run_info["UNIT"] = analyzer.analysis( run_info["FINAL_RESULT"], run_info["UNIT"] = analyzer.analysis(
batch_size=args.base_batch_size, batch_size=args.base_batch_size,
gpu_num=args.gpu_num, gpu_num=args.gpu_num,
...@@ -245,29 +297,50 @@ if __name__ == "__main__": ...@@ -245,29 +297,50 @@ if __name__ == "__main__":
run_mode=args.run_mode, run_mode=args.run_mode,
unit=args.ips_unit) unit=args.ips_unit)
try: try:
if int(os.getenv('job_fail_flag')) == 1 or int(run_info["FINAL_RESULT"]) == 0: if int(os.getenv('job_fail_flag')) == 1 or int(run_info[
"FINAL_RESULT"]) == 0:
run_info["JOB_FAIL_FLAG"] = 1 run_info["JOB_FAIL_FLAG"] = 1
except: except:
pass pass
elif args.index == 3: elif args.index == 3:
run_info["FINAL_RESULT"] = {} run_info["FINAL_RESULT"] = {}
records_fo_total = TimeAnalyzer(args.filename, 'Framework overhead', None, 3, '').records records_fo_total = TimeAnalyzer(args.filename, 'Framework overhead',
records_fo_ratio = TimeAnalyzer(args.filename, 'Framework overhead', None, 5).records None, 3, '').records
records_ct_total = TimeAnalyzer(args.filename, 'Computation time', None, 3, '').records records_fo_ratio = TimeAnalyzer(args.filename, 'Framework overhead',
records_gm_total = TimeAnalyzer(args.filename, 'GpuMemcpy Calls', None, 4, '').records None, 5).records
records_gm_ratio = TimeAnalyzer(args.filename, 'GpuMemcpy Calls', None, 6).records records_ct_total = TimeAnalyzer(args.filename, 'Computation time',
records_gmas_total = TimeAnalyzer(args.filename, 'GpuMemcpyAsync Calls', None, 4, '').records None, 3, '').records
records_gms_total = TimeAnalyzer(args.filename, 'GpuMemcpySync Calls', None, 4, '').records records_gm_total = TimeAnalyzer(args.filename,
run_info["FINAL_RESULT"]["Framework_Total"] = records_fo_total[0] if records_fo_total else 0 'GpuMemcpy Calls',
run_info["FINAL_RESULT"]["Framework_Ratio"] = records_fo_ratio[0] if records_fo_ratio else 0 None, 4, '').records
run_info["FINAL_RESULT"]["ComputationTime_Total"] = records_ct_total[0] if records_ct_total else 0 records_gm_ratio = TimeAnalyzer(args.filename,
run_info["FINAL_RESULT"]["GpuMemcpy_Total"] = records_gm_total[0] if records_gm_total else 0 'GpuMemcpy Calls',
run_info["FINAL_RESULT"]["GpuMemcpy_Ratio"] = records_gm_ratio[0] if records_gm_ratio else 0 None, 6).records
run_info["FINAL_RESULT"]["GpuMemcpyAsync_Total"] = records_gmas_total[0] if records_gmas_total else 0 records_gmas_total = TimeAnalyzer(args.filename,
run_info["FINAL_RESULT"]["GpuMemcpySync_Total"] = records_gms_total[0] if records_gms_total else 0 'GpuMemcpyAsync Calls',
None, 4, '').records
records_gms_total = TimeAnalyzer(args.filename,
'GpuMemcpySync Calls',
None, 4, '').records
run_info["FINAL_RESULT"]["Framework_Total"] = records_fo_total[
0] if records_fo_total else 0
run_info["FINAL_RESULT"]["Framework_Ratio"] = records_fo_ratio[
0] if records_fo_ratio else 0
run_info["FINAL_RESULT"][
"ComputationTime_Total"] = records_ct_total[
0] if records_ct_total else 0
run_info["FINAL_RESULT"]["GpuMemcpy_Total"] = records_gm_total[
0] if records_gm_total else 0
run_info["FINAL_RESULT"]["GpuMemcpy_Ratio"] = records_gm_ratio[
0] if records_gm_ratio else 0
run_info["FINAL_RESULT"][
"GpuMemcpyAsync_Total"] = records_gmas_total[
0] if records_gmas_total else 0
run_info["FINAL_RESULT"]["GpuMemcpySync_Total"] = records_gms_total[
0] if records_gms_total else 0
else: else:
print("Not support!") print("Not support!")
except Exception: except Exception:
traceback.print_exc() traceback.print_exc()
print("{}".format(json.dumps(run_info))) # it's required, for the log file path insert to the database print("{}".format(json.dumps(run_info))
) # it's required, for the log file path insert to the database
benchmark/run_benchmark_det.sh
浏览文件 @ 9c0a4d9d
...@@ -58,3 +58,4 @@ source ${BENCHMARK_ROOT}/scripts/run_model.sh # 在该脚本中会对符合 ...@@ -58,3 +58,4 @@ source ${BENCHMARK_ROOT}/scripts/run_model.sh # 在该脚本中会对符合
_set_params $@ _set_params $@
#_train # 如果只想产出训练log,不解析,可取消注释 #_train # 如果只想产出训练log,不解析,可取消注释
_run # 该函数在run_model.sh中,执行时会调用_train; 如果不联调只想要产出训练log可以注掉本行,提交时需打开 _run # 该函数在run_model.sh中,执行时会调用_train; 如果不联调只想要产出训练log可以注掉本行,提交时需打开
benchmark/run_det.sh
浏览文件 @ 9c0a4d9d
...@@ -36,3 +36,4 @@ for model_mode in ${model_mode_list[@]}; do ...@@ -36,3 +36,4 @@ for model_mode in ${model_mode_list[@]}; do
done done
test_tipc/benchmark_train.sh
浏览文件 @ 9c0a4d9d
...@@ -3,8 +3,6 @@ source test_tipc/common_func.sh ...@@ -3,8 +3,6 @@ source test_tipc/common_func.sh
# set env # set env
python=python python=python
export model_branch=`git symbolic-ref HEAD 2>/dev/null | cut -d"/" -f 3`
export model_commit=$(git log|head -n1|awk '{print $2}')
export str_tmp=$(echo `pip list|grep paddlepaddle-gpu|awk -F ' ' '{print $2}'`) export str_tmp=$(echo `pip list|grep paddlepaddle-gpu|awk -F ' ' '{print $2}'`)
export frame_version=${str_tmp%%.post*} export frame_version=${str_tmp%%.post*}
export frame_commit=$(echo `${python} -c "import paddle;print(paddle.version.commit)"`) export frame_commit=$(echo `${python} -c "import paddle;print(paddle.version.commit)"`)
......
test_tipc/prepare.sh
浏览文件 @ 9c0a4d9d
...@@ -24,7 +24,17 @@ if [ ${MODE} = "benchmark_train" ];then ...@@ -24,7 +24,17 @@ if [ ${MODE} = "benchmark_train" ];then
pip install -r requirements.txt pip install -r requirements.txt
if [[ ${model_name} =~ "det_mv3_db_v2_0" || ${model_name} =~ "det_r50_vd_east_v2_0" || ${model_name} =~ "det_r50_vd_pse_v2_0" || ${model_name} =~ "det_r18_db_v2_0" ]];then if [[ ${model_name} =~ "det_mv3_db_v2_0" || ${model_name} =~ "det_r50_vd_east_v2_0" || ${model_name} =~ "det_r50_vd_pse_v2_0" || ${model_name} =~ "det_r18_db_v2_0" ]];then
rm -rf ./train_data/icdar2015 rm -rf ./train_data/icdar2015
wget -nc -P ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/MobileNetV3_large_x0_5_pretrained.pdparams --no-check-certificate wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pretrained/MobileNetV3_large_x0_5_pretrained.pdparams --no-check-certificate
wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015.tar --no-check-certificate
cd ./train_data/ && tar xf icdar2015.tar && cd ../
fi
if [[ ${model_name} =~ "det_r50_vd_east_v2_0" || ${model_name} =~ "det_r50_vd_pse_v2_0" ]];then
wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pretrained/ResNet50_vd_ssld_pretrained.pdparams --no-check-certificate
wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015.tar --no-check-certificate
cd ./train_data/ && tar xf icdar2015.tar && cd ../
fi
if [[ ${model_name} =~ "det_r18_db_v2_0" ]];then
wget -nc -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pretrained/ResNet18_vd_pretrained.pdparams --no-check-certificate
wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015.tar --no-check-certificate wget -nc -P ./train_data/ https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015.tar --no-check-certificate
cd ./train_data/ && tar xf icdar2015.tar && cd ../ cd ./train_data/ && tar xf icdar2015.tar && cd ../
fi fi
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或