在DigitalOcean服务器中消除PyTorch的NNPACK警告方案求助

问题背景 #

我开发了一款基于Python的二维码识别脚本，使用依赖PyTorch的qrdet库。在DigitalOcean Linux服务器运行时，出现以下警告：

[W NNPACK.cpp:64] Could not initialize NNPACK! Reason: Unsupported hardware.

该警告不影响功能，但会破坏其他进程依赖的合法JSON输出。已尝试6种无效方案：

• 运行脚本前执行export USE_NNPACK=0设置环境变量
• 安装PyTorch前设置USE_NNPACK=0
• 运行脚本前设置CUDA_VISIBLE_DEVICES=""
• 通过pip3 install torch torchvision --index-url https://download.pytorch.org/whl/cpu重装PyTorch
• 在脚本内配置PyTorch：

x = torch.tensor([1, 2, 3])
x = x.to('cpu')
torch.backends.cudnn.enabled = False
torch.cuda.is_available = lambda : False 

• 使用warning包移除警告

环境版本信息 #

PyTorch version: 2.0.1+cpu
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A

OS: Ubuntu 18.04.6 LTS (x86_64)
GCC version: (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
Clang version: Could not collect
CMake version: version 3.27.4
Libc version: glibc-2.27

Python version: 3.8.0 (default, Dec  9 2021, 17:53:27)  [GCC 8.4.0] (64-bit runtime)
Python platform: Linux-4.15.0-213-generic-x86_64-with-glibc2.27
Is CUDA available: False
CUDA runtime version: No CUDA
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              8
On-line CPU(s) list: 0-7
Thread(s) per core:  1
Core(s) per socket:  8
Socket(s):           1
NUMA node(s):        1
Vendor ID:           GenuineIntel
CPU family:          6
Model:               63
Model name:          DO-Regular
Stepping:            2
CPU MHz:             2199.996
BogoMIPS:            4399.99
Virtualization:      VT-x
Hypervisor vendor:   KVM
Virtualization type: full
L1d cache:           32K
L1i cache:           32K
L2 cache:            4096K
NUMA node0 CPU(s):   0-7
Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx rdtscp lm constant_tsc rep_good nopl xtopology cpuid tsc_known_freq pni pclmulqdq vmx ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm cpuid_fault invpcid_single pti ssbd ibrs ibpb tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt arat md_clear

Versions of relevant libraries:
[pip3] numpy==1.24.4
[pip3] torch==2.0.1+cpu
[pip3] torchvision==0.15.2+cpu
[conda] Could not collect

可行解决方案 #

方案1：直接禁用PyTorch的NNPACK后端 #

在脚本最开头添加以下代码，强制PyTorch不使用NNPACK：

import torch
# 禁用NNPACK后端
torch.backends.nnpack.enabled = False
# 可选：若不需要XNNPACK也可一并禁用
torch.backends.xnnpack.enabled = False

方案2：过滤PyTorch底层C++警告 #

由于该警告来自PyTorch的C++层，Python的warnings模块无法捕获，可通过重定向stderr过滤特定警告：

import sys

original_stderr = sys.stderr

class FilteredStderr:
    def write(self, line):
        if "Could not initialize NNPACK! Reason: Unsupported hardware" not in line:
            original_stderr.write(line)
    def flush(self):
        original_stderr.flush()

sys.stderr = FilteredStderr()

# 必须在重定向后导入torch和qrdet
import torch
import qrdet

# 后续业务代码...

方案3：降级PyTorch版本 #

PyTorch 2.0.x在部分虚拟化CPU环境下的NNPACK检测逻辑存在问题，降级到1.13.1+cpu版本可规避：

pip3 uninstall -y torch torchvision
pip3 install torch==1.13.1+cpu torchvision==0.14.1+cpu --index-url https://download.pytorch.org/whl/cpu

方案4：全局环境变量禁用NNPACK #

在启动脚本的命令前添加环境变量，确保PyTorch加载时直接禁用NNPACK：

export TORCH_NNPACK_ENABLED=0 && python your_script.py

内容的提问来源于stack exchange，提问作者Nicolás Rossi