apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-21 15:49:31 +08:00
Code Issues Actions 6 Packages Projects Releases Wiki Activity

Add Benchmark readme (#236)

Browse Source 
* add ppcls benchmark

* add ppcls benchmark

* add ppcls benchmark

* add ppcls benchmark

* fixed txt path

* resolve conflict

* resolve conflict

* deal with comments

* Add enable_trt_fp16 option

* Add OV backend for seg and det

* fixed valid backends in ppdet

* fixed valid backends in yolo

* add copyright and rm Chinese Notes

* add ppdet&ppseg&yolo benchmark

* add cpu/gpu mem info

* Add benchmark readme

* fixed bug

Co-authored-by: Jason <jiangjiajun@baidu.com>
This commit is contained in:
WJJ1995
2022-09-15 21:36:10 +08:00
committed by GitHub
parent e9fb6141cf
commit 4d2fbcb030
5 changed files with 147 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
benchmark/benchmark_ppcls.py
View File

@@ -117,7 +117,9 @@ if __name__ == '__main__':
gpu_id = args.device_id
end2end_statis = list()
cpu_mem, gpu_mem, gpu_util = 0, 0, 0
cpu_mem = list()
gpu_mem = list()
gpu_util = list()
if args.device == "cpu":
file_path = args.model + "_model_" + args.backend + "_" + \
args.device + "_" + str(args.cpu_num_thread) + ".txt"
@@ -139,23 +141,26 @@ if __name__ == '__main__':
start = time.time()
result = model.predict(im)
end2end_statis.append(time.time() - start)
gpu_util += get_current_gputil(gpu_id)
gpu_util.append(get_current_gputil(gpu_id))
cm, gm = get_current_memory_mb(gpu_id)
cpu_mem += cm
gpu_mem += gm
cpu_mem.append(cm)
gpu_mem.append(gm)
runtime_statis = model.print_statis_info_of_runtime()
warmup_iter = args.iter_num // 5
repeat_iter = args.iter_num - warmup_iter
end2end_statis = end2end_statis[warmup_iter:]
end2end_statis_repeat = end2end_statis[warmup_iter:]
cpu_mem_repeat = cpu_mem[warmup_iter:]
gpu_mem_repeat = gpu_mem[warmup_iter:]
gpu_util_repeat = gpu_util[warmup_iter:]
dump_result = dict()
dump_result["runtime"] = runtime_statis["avg_time"] * 1000
dump_result["end2end"] = np.mean(end2end_statis) * 1000
dump_result["cpu_rss_mb"] = cpu_mem / repeat_iter
dump_result["gpu_rss_mb"] = gpu_mem / repeat_iter
dump_result["gpu_util"] = gpu_util / repeat_iter
dump_result["end2end"] = np.mean(end2end_statis_repeat) * 1000
dump_result["cpu_rss_mb"] = np.mean(cpu_mem_repeat)
dump_result["gpu_rss_mb"] = np.mean(gpu_mem_repeat)
dump_result["gpu_util"] = np.mean(gpu_util_repeat)
f.writelines("Runtime(ms): {} \n".format(str(dump_result["runtime"])))
f.writelines("End2End(ms): {} \n".format(str(dump_result["end2end"])))