pytorch_quantization报错

CUDA error: no kernel image is available for execution on the device

但是相同环境下去掉pytorch_quantization中的init, 不实用QAT即可成功运行

原因是pytorch_quantization安装的不符合当前GPU架构需要重新安装pytorch_quantization【源码安装】

https://github.com/NVIDIA/TensorRT/issues/3497

git clone https://github.com/NVIDIA/TensorRT.git
cd tools/pytorch-quantization
安装 PyTorch 和先决条件

pip install -r requirements.txt
# for CUDA 10.2 users
pip install torch>=1.9.1
# for CUDA 11.1 users
pip install torch>=1.9.1+cu111
构建并安装 pytorch-quantization

# Python version >= 3.7, GCC version >= 5.4 required
python setup.py install