RTX 3050显卡配置yolov5_DeepSort_Traffic-counter-main运行环境

终端中运行程序报错:

NVIDIA GeForce RTX 3050 with CUDA capability sm_86 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70.
If you want to use the NVIDIA GeForce RTX 3050 GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/

安装匹配cuda11.3版本的pytorch

运行如下命令：
conda install pytorch1.12.1 torchvision0.13.1 torchaudio==0.12.1 cudatoolkit=11.3 -c pytorch

把requirements.txt内容替换成如下

# pip install -r requirements.txt

# base ----------------------------------------

matplotlib>=3.2.2
numpy==1.23.5
opencv-python>=4.1.2
Pillow
PyYAML>=5.3.1
scipy>=1.4.1
# torch==1.9.1
# torchvision==0.10.1
tqdm>=4.41.0

# plotting ------------------------------------

seaborn>=0.11.0
pandas

# deep_sort -----------------------------------

easydict
requests

报错:

raise AttributeError("'{}' object has no attribute '{}'".format(

AttributeError: ‘Upsample’ object has no attribute ‘recompute_scale_factor’

解决方法：
https://blog.csdn.net/Thebest_jack/article/details/124723687