Getting "Could not initialize NNPACK! Reason: Unsupported hardware." warning even though NNPACK is enabled

[kirillmeisser]

Hi everyone,

I am trying to deploy EasyOCR locally on a VM and when executing the output = reader.readtext(image_array) command I get the following warning: "Could not initialize NNPACK! Reason: Unsupported hardware.". I am deploying in a CPU only environment, on CPUs with the AVX512 instructions enabled. When the warning is displayed the model takes a lot more time to process and triggers a Timeout. I executed the following command print(torch.__config__.show()) to see if NNPACK is available at runtime and indeed it is. This is the output right before the inference is processed:

PyTorch built with:
    - GCC 9.3
    - C++ Version: 201703
    - Intel(R) oneAPI Math Kernel Library Version 2022.2-Product Build 20220804 for Intel(R) 64 architecture applications
    - Intel(R) MKL-DNN v3.4.2 (Git Hash 1137e04ec0b5251ca2b4400a4fd3c667ce843d67)
    - OpenMP 201511 (a.k.a. OpenMP 4.5)
    - LAPACK is enabled (usually provided by MKL)
    - NNPACK is enabled
    - CPU capability usage: AVX512
    - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 -fabi-version=11 -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOCUPTI -DLIBKINETO_NOROCTRACER -DUSE_FBGEMM -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Werror=bool-operation -Wnarrowing -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wsuggest-override -Wno-psabi -Wno-error=pedantic -Wno-error=old-style-cast -Wno-missing-braces -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=2.4.0, USE_CUDA=0, USE_CUDNN=OFF, USE_CUSPARSELT=OFF, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_GLOO=ON, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=OFF, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, USE_ROCM_KERNEL_ASSERT=OFF, 

I don't know why it does not detect NNPACK, when pytorch is built with this capability. Any help would be greatly appreciated.

My environment is:

easyocr==1.7.1
torch==2.4.0
torchvision==0.19.0