Armbian安装numpy、scipy的另类方法

直接安装numpy、scipy会出现

libraries mkl_rt not found
libraries blis not found等库缺失的情况，具体要再安装哪个库确实难找，可以安装python3-keras来附带安装numpy和scipy

apt install python3-keras

apt install python3-keras
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
The following additional packages will be installed:
  libaec0 libhdf5-103-1 libhdf5-hl-100 liblbfgsb0 liblzf1 libsz2 python3-decorator python3-h5py python3-h5py-serial python3-keras-applications
  python3-keras-preprocessing python3-numpy python3-scipy python3-six python3-theano
Suggested packages:
  python-h5py-doc python-numpy-doc python3-numpy-dbg python3-pytest python-scipy-doc nvidia-cuda-toolkit python3-pycuda theano-doc
Recommended packages:
  python3-pil python3-pydot python3-nose python3-parameterized python3-pygpu libgpuarray-dev cython3 graphviz
The following NEW packages will be installed:
  libaec0 libhdf5-103-1 libhdf5-hl-100 liblbfgsb0 liblzf1 libsz2 python3-decorator python3-h5py python3-h5py-serial python3-keras python3-keras-applications
  python3-keras-preprocessing python3-numpy python3-scipy python3-six python3-theano
0 upgraded, 16 newly installed, 0 to remove and 1 not upgraded.
Need to get 18.6 MB of archives.
After this operation, 76.2 MB of additional disk space will be used.

可以看出，附带安装了python3-numpy python3-scipy，估计也可以使用如下独立安装

apt install python3-numpy

apt install python3-scipy

注意安装了python3-keras后，mkl_rt 、blis 依然是没有的