# Benchmarks Targeted for Jetson TX2

**URL:** <https://forums.developer.nvidia.com/t/benchmarks-targeted-for-jetson-tx2/232954>\
**Category:** Jetson TX2\
**Tags:** tensorrt, jetson-inference\
**Created:** [November 6, 2022, 6:11am UTC](https://forums.developer.nvidia.com/t/benchmarks-targeted-for-jetson-tx2/232954 "2022-11-06T06:11:38Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![jslim0326](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@jslim0326](https://forums.developer.nvidia.com/u/jslim0326)\
**Post date:** [November 6, 2022, 6:11am UTC](https://forums.developer.nvidia.com/t/benchmarks-targeted-for-jetson-tx2/232954/1 "2022-11-06T06:11:38Z")

</div>

Hello,

I’m studying about ‘the Autonomous landing of small UAV’ using jetson nano.  
I built using SSD MobileNet v2 for it and done it.  
But FPS is too low as 0.5~0.8 using SSD-MobileNet v2.  
So I’ve changed the board to TX2 but it’s same.  
Then, I’ve tried to using TensorRT, Benchmars etc.

I’ve just found the git-hub as below,  
[GitHub - NVIDIA-AI-IOT/jetson\_benchmarks: Jetson Benchmark](https://github.com/NVIDIA-AI-IOT/jetson_benchmarks).

But I got the error msg as below, but I have no idea how to solve it.  
Please help me.

============error=========================

kookmin-uav@ubuntu:~/jetson\_benchmarks$ sudo python3 benchmark.py --all --csv\_file\_path /home/kookmin-uav/jetson\_benchmarks/benchmark\_csv/tx2-nano-benchmarks.csv \

> ```
> --model_dir /home/kookmin-uav/jetson_benchmarks/models \
> --jetson_devkit tx2 \
> --gpu_freq 1122000000 --power_mode 3 --precision fp16
> 
> ```

Please close all other applications and Press Enter to continue…  
Setting Jetson tx2 in max performance mode  
Traceback (most recent call last):  
File “benchmark.py”, line 130, in   
main()  
File “benchmark.py”, line 28, in main  
system\_check.run\_set\_clocks\_withDVFS()  
File “/home/kookmin-uav/jetson\_benchmarks/utils/utilities.py”, line 34, in run\_set\_clocks\_withDVFS  
self.set\_clocks\_withDVFS(frequency=self.gpu\_freq, device=‘gpu’)  
File “/home/kookmin-uav/jetson\_benchmarks/utils/utilities.py”, line 74, in set\_clocks\_withDVFS  
from\_freq = self.read\_internal\_register(register=freq\_register\_, device=device)  
File “/home/kookmin-uav/jetson\_benchmarks/utils/utilities.py”, line 100, in read\_internal\_register  
reg\_read = open(register, “r”)  
FileNotFoundError: [Errno 2] No such file or directory: ‘/sys/devices/gpu.0/devfreq/17000000.gp10bnafll\_gpc0/rate’

---

<div class="post-metadata">

**Author:** ![AastaLLL](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aastalll/32/14043_2.png) [@AastaLLL](https://forums.developer.nvidia.com/u/AastaLLL)\
**Post date:** [November 7, 2022, 6:57am UTC](https://forums.developer.nvidia.com/t/benchmarks-targeted-for-jetson-tx2/232954/3 "2022-11-07T06:57:33Z")

</div>

Hi,

Please add `--jetson_clocks` to enable the clock with _jetson\_clocks_ instead.  
For example:

```auto
$ sudo python3 benchmark.py --all --csv_file_path /home/nvidia/topic_232954/jetson_benchmarks/benchmark_csv/tx2-nano-benchmarks.csv --jetson_clocks

```

Thanks.

---

<div class="post-metadata">

**Author:** ![jslim0326](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@jslim0326](https://forums.developer.nvidia.com/u/jslim0326)\
**Post date:** [November 8, 2022, 2:00am UTC](https://forums.developer.nvidia.com/t/benchmarks-targeted-for-jetson-tx2/232954/4 "2022-11-08T02:00:00Z")

</div>

Thank you so much your help!  
I’ve just make it run!

By the way, I have one more question.  
I’d like to ask you this benchmark can use with pre-trained model by tf2?

Actually I trained SSD-MobileNet v2 by tf2 for UAV’s autonomous landing.  
This model can detect helipad BUT FPS is appx. 0.5 with Jetson nano.  
So I would love to make it higher but have no idea.  
I’m trying to use Jetson-inference, too but I don’t know how to use my pre-trained model.  
Not using benchmark is okay, If I can make FPS higher.  
Please help me how to solve this problem.

Thank you so much!

---

<div class="post-metadata">

**Author:** ![AastaLLL](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aastalll/32/14043_2.png) [@AastaLLL](https://forums.developer.nvidia.com/u/AastaLLL)\
**Post date:** [November 17, 2022, 6:45am UTC](https://forums.developer.nvidia.com/t/benchmarks-targeted-for-jetson-tx2/232954/5 "2022-11-17T06:45:28Z")

</div>

Hi,

You can use _trtexec_ for the benchmark.

Below is a tutorial that converts a TF2 model into TensorRT for your reference:

> **[TensorRT/samples/python/tensorflow\_object\_detection\_api at release/8.2 ·...](https://github.com/NVIDIA/TensorRT/tree/release/8.2/samples/python/tensorflow_object_detection_api)**
>
> release/8.2/samples/python/tensorflow\_object\_detection\_api

Thanks.

---

<div class="post-metadata">

**Author:** ![system](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/system/32/68080_2.png) [@system](https://forums.developer.nvidia.com/u/system)\
**Post date:** [December 7, 2022, 5:19am UTC](https://forums.developer.nvidia.com/t/benchmarks-targeted-for-jetson-tx2/232954/7 "2022-12-07T05:19:57Z")

</div>

Closed on December 7, 2022, 5:19am UTC

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.