YoloV8 problems (core dumped)

[predrag87]

When running task problem happens(GPU,CPU is the same)

yolo task=detect mode=predict model=yolov8n.pt source=0

Ultralytics 8.3.107 🚀 Python-3.10.12 torch-2.3.0 CUDA:0 (Orin, 62841MiB)
YOLOv8n summary (fused): 72 layers, 3,151,904 parameters, 0 gradients, 8.7 GFLOPs

1/1: 0… Success ✅ (inf frames of shape 640x480 at 30.00 FPS)

/opt/rh/gcc-toolset-14/root/usr/include/c++/14/bits/stl_vector.h:1130: std::vector<_Tp, _Alloc>::reference std::vector<_Tp, _Alloc>::operator [with _Tp = unsigned int; _Alloc = std::allocator; reference = unsigned int&; size_type = long unsigned int]: Assertion ‘__n < this->size()’ failed.
Aborted (core dumped)

CUDA Version: 12.6
CUDA version in PyTorch: 12.4

[DavidDDD]

Hi,

What is your JP version?
How do you isntall the pytorch and yolo?
Please provide your commands for us to review.

Thanks

[karthik.a1]

Hi @DavidDDD
Iam also facing same issue

user@saartha-desktop:~$ yolo task=detect mode=predict model=yolov8n.pt source=0
Ultralytics 8.3.108 🚀 Python-3.10.12 torch-2.3.0 CUDA:0 (Orin, 15656MiB)
YOLOv8n summary (fused): 72 layers, 3,151,904 parameters, 0 gradients, 8.7 GFLOPs

1/1: 0... Success ✅ (inf frames of shape 1344x376 at 30.00 FPS)

/opt/rh/gcc-toolset-14/root/usr/include/c++/14/bits/stl_vector.h:1130: std::vector<_Tp, _Alloc>::reference std::vector<_Tp, _Alloc>::operator[](size_type) [with _Tp = unsigned int; _Alloc = std::allocator<unsigned int>; reference = unsigned int&; size_type = long unsigned int]: Assertion '__n < this->size()' failed.
Aborted (core dumped)
user@saartha-desktop:~

I am using jetpack 6.0

[DavidDDD]

Hi,

How do you install the pytorch and yolo?
Please provide your commands for us to review.

Thanks

[karthik.a1]

I followed below documentation

https://forums.developer.nvidia.com/t/pytorch-for-jetson/72048

Thank you