rpi5 example error after upgrading to 4.20

General
,
1

After upgrading the pci driver and tools to latest release 4.20, I’ve got the following error when running the rpi5 default exemple script detection.py

terminate called after throwing an instance of ‘std::invalid_argument’
what(): Output tensor yolov8m/yolov8_nms_postprocess is not an NMS type
Aborted

Any ideas ?

2

Hey @neoyoyo ,

Welcome to the Hailo Community!

This issue should be resolved , are you still facing this issue ?

3

Hi Omria,

I add to reinstall my RPI5 from scratch in order to have the system working again. I installed the 4.19 version, and didn’t try to upgrade to 4.20.

BTW, if you have a procedure in order to clean an installation that could avoid a full reinstall.

Before reinstalling, I try to recompile, and after that, I get a seg fault error…

did you fix something into the code ?

4

I’m also having problem after upgrading to 4.20.0, I use library libyolo_hailortpp_postprocess.so using python 3 I have an error:

terminate called after throwing an instance of ‘std::bad_alloc’
what(): std::bad_alloc
Aborted

Version 4.19.0 worked well, I don’t see that the library in question has been changed.

1 Like
5

Same here, I am also experiencing what @Toni_Stojev reports, even after applying the recent patch.

6

Hey @eren and @Toni_Stojev,

Could you please provide more details about this issue:

  • When does this error occur?
  • Which example were you running when you encountered it?
  • Could you share the complete log output or error message?

This information will help us better understand and address the problem.

7

The problem is probably related to the GStreamerApp, i had an app based on an example using it.
That example reports Segmentation fault after upgrade to 4.20.0.
Example with GStreamerDetectionAplication works well, so I have moved to using that instead of GStreamerApp and I don’t have problems now.

8

Hello,
Also receiving this error.

ony@raspberrypi:~/hailo-rpi5-examples $ source ./setup_env.sh
Setting up the environment…
Setting up the environment for hailo-tappas-core…
TAPPAS_VERSION is 3.31.0. Proceeding…
You are not in the venv_hailo_rpi5_examples virtual environment.
Virtual environment exists. Activating…
TAPPAS_POST_PROC_DIR set to /usr/lib/aarch64-linux-gnu/hailo/tappas/post_processes
DEVICE_ARCHITECTURE is set to: HAILO8
(venv_hailo_rpi5_examples) tony@raspberrypi:~/hailo-rpi5-examples $ python basic_pipelines/detection.py --input resources/detection0.mp4
Auto-detected Hailo architecture: hailo8
filesrc location=“resources/detection0.mp4” name=source ! queue name=source_queue_dec264 leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! qtdemux ! h264parse ! avdec_h264 max-threads=2 ! queue name=source_scale_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! videoscale name=source_videoscale n-threads=2 ! queue name=source_convert_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! videoconvert n-threads=3 name=source_convert qos=false ! video/x-raw, format=RGB, pixel-aspect-ratio=1/1 ! queue name=inference_scale_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! videoscale name=inference_videoscale n-threads=2 qos=false ! queue name=inference_convert_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! video/x-raw, pixel-aspect-ratio=1/1 ! videoconvert name=inference_videoconvert n-threads=2 ! queue name=inference_hailonet_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! hailonet name=inference_hailonet hef-path=/home/tony/hailo-rpi5-examples/basic_pipelines/…/resources/yolov8m.hef batch-size=2 nms-score-threshold=0.3 nms-iou-threshold=0.45 output-format-type=HAILO_FORMAT_TYPE_FLOAT32 force-writable=true ! queue name=inference_hailofilter_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! hailofilter name=inference_hailofilter so-path=/home/tony/hailo-rpi5-examples/basic_pipelines/…/resources/libyolo_hailortpp_postprocess.so qos=false ! queue name=identity_callback_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! identity name=identity_callback ! queue name=hailo_display_hailooverlay_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! hailooverlay name=hailo_display_hailooverlay ! queue name=hailo_display_videoconvert_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! videoconvert name=hailo_display_videoconvert n-threads=2 qos=false ! queue name=hailo_display_q leaky=no max-size-buffers=3 max-size-bytes=0 max-size-time=0 ! fpsdisplaysink name=hailo_display video-sink=xvimagesink sync=true text-overlay=false signal-fps-measurements=true
Showing FPS
terminate called after throwing an instance of ‘std::invalid_argument’
what(): Output tensor yolov8m/yolov8_nms_postprocess is not an NMS type
Aborted

Thoughts?

9

Same problem here… did you get it solved? … Output tensor yolov8m/yolov8_nms_postprocess is not an NMS type
Aborted

10

I couldn’t fully work it out. I did a full rebuild which then caused it to work first time.
Would still be interested to know what I did wrong myself on that attempt.

11

I am running a custom python file that contains object detection using gstreamer pipeline. When I am executing the file, I am facing
std::bad-alloc error. I do not know what to do. Can anyone please help me?

(venv_hailo_rpi5_examples) spectrum@raspberrypi:~/Desktop/Himanshu/hailo-rpi5-examples/ms_temp $ python 20-03-2025-jsonl.py
Creating GStreamer Pipeline…
Pipeline running… Press Ctrl+C to stop.
terminate called after throwing an instance of ‘std::bad_alloc’
what(): std::bad_alloc
Aborted
(venv_hailo_rpi5_examples) spectrum@raspberrypi:~/Desktop/Himanshu/hailo-rpi5-examples/ms_temp $