Not able to run benchmark (getting 0 FPS)

jasvinderkhurana commented 4 years ago

Hi

I am trying to run the jetson benchmarks, but when I execute it, I am getting 0 FPS, and seems not even a single model is running

` Please close all other applications and Press Enter to continue... Setting Jetson nano in max performance mode gpu frequency is set from 921600000 Hz --> to 921600000 Hz Running all benchmarks.. This will take at least 2 hours... ------------Executing inception_v4------------

Error in Build, Please check the log in: ./models_nano We recommend to run benchmarking in headless mode

Model Name: inception_v4 FPS:0.00

------------Executing vgg19_N2------------

Error in Build, Please check the log in: ./models_nano We recommend to run benchmarking in headless mode

Model Name: vgg19_N2 FPS:0.00

------------Executing super_resolution_bsd500------------

Error in Build, Please check the log in: ./models_nano We recommend to run benchmarking in headless mode

Model Name: super_resolution_bsd500 FPS:0.00

------------Executing unet-segmentation------------

Error in Build, Please check the log in: ./models_nano We recommend to run benchmarking in headless mode

Model Name: unet-segmentation FPS:0.00

------------Executing pose_estimation------------

Error in Build, Please check the log in: ./models_nano We recommend to run benchmarking in headless mode

Model Name: pose_estimation FPS:0.00

------------Executing yolov3-tiny-416------------

Error in Build, Please check the log in: ./models_nano We recommend to run benchmarking in headless mode

Model Name: yolov3-tiny-416 FPS:0.00

------------Executing ResNet50_224x224------------

Error in Build, Please check the log in: ./models_nano We recommend to run benchmarking in headless mode

Model Name: ResNet50_224x224 FPS:0.00

------------Executing ssd-mobilenet-v1------------

Error in Build, Please check the log in: ./models_nano We recommend to run benchmarking in headless mode

Model Name: ssd-mobilenet-v1 FPS:0.00

            Model Name  FPS

0 inception_v4 0.0 1 vgg19_N2 0.0 2 super_resolution_bsd500 0.0 3 unet-segmentation 0.0 4 pose_estimation 0.0 5 yolov3-tiny-416 0.0 6 ResNet50_224x224 0.0 7 ssd-mobilenet-v1 0.0 `

Can anyone guide what is the problem here?

jasvinderkhurana commented 4 years ago

inception_v4_b1_ws1024_gpu.txt

Logs attached

ak-nv commented 4 years ago

You need to provide absolute path to models directory as mentioned in readme.

GabrielLima1995 commented 3 years ago

@jasvinderkhurana did you solve this problem ? I'm getting the same issue and i'm providing the absolute path .

mohammadhaqqi commented 3 years ago

same proble file not found in models directory Error opening engine file: models/ResNet50_224x224_b4_ws2048_gpu.engine no file *.engine found ???

GabrielLima1995 commented 3 years ago

@mohammadhaqqi I have found the answer for this problem here, and for me the second step has worked , so I only decreased the workspace size in this file as mentioned in previous link.

rickymedrano commented 3 years ago

@jasvinderkhurana @mohammadhaqqi I was having the same issue, even when running headless. Turns out I left off the "fp" in "fp16" in the call. Here is what I used to get it work: sudo python3 benchmark.py --all --csv_file_path <path-to>/benchmark_csv/tx2-nano-benchmarks.csv \ --model_dir /home/nvidia/Desktop/jetson_benchmarks/models \ --jetson_devkit nano \ --gpu_freq 921600000 --power_mode 0 --precision fp16

yueyihua commented 3 years ago

Can you put the model on NVIDIA website or Google Drive？I can't download them from dropbox.

rickymedrano commented 3 years ago

@yueyihua you can get individual download links from here https://github.com/NVIDIA-AI-IOT/jetson_benchmarks/blob/master/benchmark_csv/xavier-benchmarks.csv

yueyihua commented 3 years ago

@rickymedrano I see all of the links are from dropbox. ModelName FrameWork Devices BatchSizeGPU BatchSizeDLA WS_GPU WS_DLA input output URL inception_v4 caffe 3 4 1 2048 2048 NA prob https://www.dropbox.com/s/b7masj8xdoycv2w/inception_v4.prototxt vgg19_N2 caffe 1 4 0 2048 0 NA prob https://www.dropbox.com/s/t4qq079g5q4jibx/vgg19_N2.prototxt super_resolution_bsd500 onnx 1 4 0 2048 2048 NA NA https://www.dropbox.com/s/hdhxndo23cm9i5y/super_resolution_bsd500.zip unet-segmentation tensorrt 1 2 0 2048 None input_1,1,512,512 conv2d_19/Sigmoid https://www.dropbox.com/s/85lttamnbjeig0e/unet-segmentation.uff pose_estimation caffe 1 4 0 2048 None NA Mconv7_stage2_L2 https://www.dropbox.com/s/hwa5i14v67u57ij/pose_estimation.prototxt yolov3-tiny-416 onnx 1 16 0 2048 2048 NA NA https://www.dropbox.com/s/ck9e40b57rd5o14/yolov3-tiny-416.zip ResNet50_224x224 caffe 3 16 4 2048 2048 NA prob https://www.dropbox.com/s/9ohk387v0ki56wx/ResNet50_224x224.prototxt ssd-mobilenet-v1 onnx 3 16 2 2048 2048 NA NA https://www.dropbox.com/s/gx5zayt76vszhpo/ssd-mobilenet-v1.zip

rickymedrano commented 3 years ago

@yueyihua What's the issue? I clicked on the links and they work.

yueyihua commented 3 years ago

Sorry, In China, Dropbox is not accessible!

---Original--- From: "Ricky Medrano"<notifications@github.com> Date: Sat, Mar 6, 2021 00:51 AM To: "NVIDIA-AI-IOT/jetson_benchmarks"<jetson_benchmarks@noreply.github.com>; Cc: "yueyihua"<david.yue@qq.com>;"Mention"<mention@noreply.github.com>; Subject: Re: [NVIDIA-AI-IOT/jetson_benchmarks] Not able to run benchmark (getting 0 FPS) (#11)

@yueyihua What's the issue? I clicked on the links and they work.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

ak-nv commented 3 years ago

Let me figure out other options.

Chakib08 commented 3 years ago

Add $PWD before models in the command line like that :

sudo python3 benchmark.py --all --csv_file_path benchmark_csv/xavier-benchmarks.csv --model_dir $PWD/models --jetson_devkit xavier --gpu_freq 1377000000 --dla_freq 1395200000 --power_mode 0 It worked in my case

NVIDIA-AI-IOT / jetson_benchmarks