After Jetson Orin Nano upgraded Super mode,CUDA12.6+Pytorch2.6.0+cu126
系统信息：
Python版本: 3.9.21 (main, Dec 11 2024, 16:27:47)
[GCC 11.2.0]
NumPy版本: 1.26.4
NVIDIA驱动信息：
Thu Feb 13 09:29:06 2025
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 540.4.0 Driver Version: 540.4.0 CUDA Version: 12.6 |
|-----------------------------------------+----------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 Orin (nvgpu) N/A | N/A N/A | N/A |
| N/A N/A N/A N/A / N/A | Not Supported | N/A N/A |
| | | N/A |
+-----------------------------------------+----------------------+----------------------+
+---------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| No running processes found |
+---------------------------------------------------------------------------------------+
CUDA版本信息：
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Tue_Oct_29_23:53:06_PDT_2024
Cuda compilation tools, release 12.6, V12.6.85
Build cuda_12.6.r12.6/compiler.35059454_0
检测到CUDA版本: 12.6
CUDNN信息：
在 /usr/local/cuda/lib64 中找到CUDNN库:
-rw-r--r-- 1 root root 137008 2月 12 17:24 /usr/local/cuda/lib64/libcudnn.so
-rw-r--r-- 1 root root 137008 2月 12 17:24 /usr/local/cuda/lib64/libcudnn.so.9
-rw-r--r-- 1 root root 137008 2月 12 17:24 /usr/local/cuda/lib64/libcudnn.so.9.5.1
在 /usr/local/cuda/targets/aarch64-linux/lib 中找到CUDNN库:
-rw-r--r-- 1 root root 137008 2月 12 17:24 /usr/local/cuda/targets/aarch64-linux/lib/libcudnn.so
-rw-r--r-- 1 root root 137008 2月 12 17:24 /usr/local/cuda/targets/aarch64-linux/lib/libcudnn.so.9
-rw-r--r-- 1 root root 137008 2月 12 17:24 /usr/local/cuda/targets/aarch64-linux/lib/libcudnn.so.9.5.1
系统架构：
aarch64
库路径信息：
LD_LIBRARY_PATH: /usr/local/cuda-12.6/lib64:/usr/local/cuda-12.6/lib64:
Python包信息：
torch 2.6.0+cu126
torchaudio 2.6.0
torchvision 0.21.0
检查PyTorch安装...
PyTorch版本: 2.6.0+cu126
CUDA是否可用: True
CUDA版本: 12.6
cuDNN版本: 90501
GPU设备: Orin
GPU内存信息：
总显存: 3.52 GB
当前占用: 0.00 GB
当前缓存: 0.00 GB
CUDA测试成功！
When training the Yolo v5 7.0 model, the following error occurred
Traceback (most recent call last):
File "/home/user/Desktop/yolov5-7.0/train.py", line 635, in <module>
main(opt)
File "/home/user/Desktop/yolov5-7.0/train.py", line 529, in main
train(opt.hyp, opt, device, callbacks)
File "/home/user/Desktop/yolov5-7.0/train.py", line 126, in train
model = Model(cfg or ckpt['model'].yaml, ch=3, nc=nc, anchors=hyp.get('anchors')).to(device) # create
File "/usr/local/miniconda3/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1343, in to
return self._apply(convert)
File "/home/user/Desktop/yolov5-7.0/models/yolo.py", line 155, in _apply
self = super()._apply(fn)
File "/usr/local/miniconda3/lib/python3.9/site-packages/torch/nn/modules/module.py", line 903, in _apply
module._apply(fn)
File "/usr/local/miniconda3/lib/python3.9/site-packages/torch/nn/modules/module.py", line 903, in _apply
module._apply(fn)
File "/usr/local/miniconda3/lib/python3.9/site-packages/torch/nn/modules/module.py", line 903, in _apply
module._apply(fn)
File "/usr/local/miniconda3/lib/python3.9/site-packages/torch/nn/modules/module.py", line 930, in _apply
param_applied = fn(param)
File "/usr/local/miniconda3/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1329, in convert
return t.to(
RuntimeError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.