picodet模型在3060gpu上 fps只有 average latency time(ms): 77.04, QPS: 12.980700 只有１２帧？？？ 大家能跑到官网说的120帧率嘛

[pureloveljc]

PaddleDetection team appreciate any suggestion or problem you delivered~

Checklist:

1. 查找历史相关issue寻求解答/I have searched related issues but cannot get the expected help.
2. 翻阅FAQ /I have read the FAQ documentation but cannot get the expected help.

描述问题/Describe the bug

A clear and concise description of what the bug is.

复现/Reproduction

1. 您使用的命令是？/What command or script did you run?

请填写命令/A placeholder for the command.

2. 您是否更改过代码或配置文件？您是否理解您所更改的内容？还请您提供所更改的部分代码。/Did you make any modifications on the code or config? Did you understand what you have modified? Please provide the codes that you modified.

3. 您使用的数据集是？/What dataset did you use?

4. 请提供您出现的报错信息及相关log。/Please provide the error messages or relevant log information.

环境/Environment

1. 请提供您使用的Paddle和PaddleDetection的版本号/Please provide the version of Paddle and PaddleDetection you use：

2. 如您在使用PaddleDetection的同时还在使用其他产品，如PaddleServing、PaddleInference等，请您提供其版本号/ Please provide the version of any other related tools/products used, such as the version of PaddleServing and etc：

3. 请提供您使用的操作系统信息，如Linux/Windows/MacOS /Please provide the OS information, e.g., Linux：

4. 请问您使用的Python版本是？/ Please provide the version of Python you used.

5. 请问您使用的CUDA/cuDNN的版本号是？/ Please provide the version of CUDA/cuDNN you used.

如果您的issue是关于安装或环境，您可以先查询安装文档尝试解决~

If your issue looks like an installation issue / environment issue, please first try to solve it yourself with the instructions in https://github.com/PaddlePaddle/PaddleDetection/blob/release/2.1/docs/tutorials/INSTALL.md

[wangxinxin08]

你好，你是怎么进行测速的呢？

[pureloveljc]

python deploy/python/infer.py --model_dir=./inference_model/picodet_s_320_voc --image_dir=./dataset/dqist_qdcode/JPEGImages/ --device=GPU 用的这个脚本

[wangxinxin08]

@pureloveljc 可以只测一张图片，然后加上--run_benchmark=True这个flag，再测一下看看

[pureloveljc]

测一张图片没有意义啊　 目前来说是这个 最多20帧 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] model_name: picodet_s_320_voc 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] precision: paddle 2022-01-24 17:30:51,727 - benchmark_utils - INFO - ----------------------- Data info ----------------------- 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] batch_size: 1 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] input_shape: dynamic_shape 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] data_num: 972 2022-01-24 17:30:51,727 - benchmark_utils - INFO - ----------------------- Perf info ----------------------- 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] cpu_rss(MB): 3449, cpu_vms: 0, cpu_shared_mb: 0, cpu_dirty_mb: 0, cpu_util: 0% 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] gpu_rss(MB): 2206, gpu_util: 17.79%, gpu_mem_util: 0% 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] total time spent(s): 58.4044 2022-01-24 17:30:51,727 - benchmark_utils - INFO - [Det] preprocess_time(ms): 50.1, inference_time(ms): 8.5, postprocess_time(ms): 1.5

[wangxinxin08]

因为picodet的目标硬件是arm cpu和intel cpu，所以官方给的fps并不是在gpu上测的，你可以按照官方教程在arm cpu上测试一下速度，另外inference time是模型在网络中运行的时间

[paddle-bot-old[bot]]

Since this issue has not been updated for more than three months, it will be closed, if it is not solved or there is a follow-up one, please reopen it at any time and we will continue to follow up. It is recommended to pull and try the latest code first. 由于该问题超过三个月未更新，将会被关闭，若问题未解决或有后续问题，请随时重新打开（建议先拉取最新代码进行尝试），我们会继续跟进。