安裝順序：
$sudo apt-get install libopenblas-base
$pip3 install mxnet-jetson --user
問題解決：
PyPI[] 安裝mxnet-jetson 依賴的是numpy 0.15.0版本但是使用時(shí)會(huì)出現(xiàn)類似錯(cuò)誤：“ImportError: No module named 'numpy.core._multiarray_umath'”
需要把numpy 更新為1.16.2 版本萌抵，即：pip3 install --upgrade numpy --user
nvidia@tegra-ubuntu:~/work$ $pip3 install mxnet-jetson --user
***********************************************************************
nvidia@tegra-ubuntu:~/work$ python3
Python 3.5.2 (default, Nov 12 2018, 13:43:14)
[GCC 5.4.0 20160609] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import mxnet
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/__init__.py", line 25, in <module>
from . import engine
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/engine.py", line 23, in <module>
from .base import _LIB, check_call
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/base.py", line 113, in <module>
_LIB = _load_lib()
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/base.py", line 105, in _load_lib
lib = ctypes.CDLL(lib_path[0], ctypes.RTLD_LOCAL)
File "/usr/lib/python3.5/ctypes/__init__.py", line 347, in __init__
self._handle = _dlopen(self._name, mode)
OSError: libopenblas.so.0: cannot open shared object file: No such file or directory
nvidia@tegra-ubuntu:~/work$ python3
Python 3.5.2 (default, Nov 12 2018, 13:43:14)
[GCC 5.4.0 20160609] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import mxnet
ImportError: No module named 'numpy.core._multiarray_umath'
ImportError: No module named 'numpy.core._multiarray_umath'
ImportError: No module named 'numpy.core._multiarray_umath'
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/__init__.py", line 28, in <module>
from . import contrib
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/contrib/__init__.py", line 31, in <module>
from . import onnx
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/contrib/onnx/__init__.py", line 20, in <module>
from ._import.import_model import import_model
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/contrib/onnx/_import/__init__.py", line 20, in <module>
from . import import_model
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/contrib/onnx/_import/import_model.py", line 22, in <module>
from .import_onnx import GraphProto
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/contrib/onnx/_import/import_onnx.py", line 22, in <module>
from .... import symbol
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/symbol/__init__.py", line 20, in <module>
from . import _internal, contrib, linalg, op, random, sparse, image
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/symbol/contrib.py", line 22, in <module>
from .random import uniform
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/symbol/random.py", line 22, in <module>
from .symbol import Symbol
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/symbol/symbol.py", line 44, in <module>
from ..executor import Executor
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/executor.py", line 36, in <module>
from .executor_manager import _split_input_slice, _check_arguments, _load_data, _load_label
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/executor_manager.py", line 29, in <module>
from .io import DataDesc
File "/home/nvidia/.local/lib/python3.5/site-packages/mxnet/io.py", line 27, in <module>
import h5py
File "/home/nvidia/.local/lib/python3.5/site-packages/h5py/__init__.py", line 36, in <module>
from ._conv import register_converters as _register_converters
File "h5py/h5t.pxd", line 14, in init h5py._conv
File "h5py/h5t.pyx", line 1, in init h5py.h5t
SystemError: execution of module h5py.utils raised unreported exception
Step1. Check depend packages
mxnet tx2版本采用的是openblas诈铛，需要安裝libopenblas-dev
$sudo apt-get update
$sudo apt-get install libopenblas-dev
$sudo apt-get -y install git build-essential libatlas-base-dev libopencv-dev graphviz python-pip
$sudo pip install pip --upgrade
$sudo pip install setuptools numpy --upgrade
$sudo pip install graphviz jupyter
Step2. Download mxnet and 3rdparty packages
由于mxnet 源碼中不包含mshadow和googletest 源碼勿璃，需要單獨(dú)clone這兩個(gè)源碼到3rdparty 目錄
$git clone https://github.com/apache/incubator-mxnet.git --recursive
$cd incubator-mxnet
$cd 3rdparty/
$git clone git clone https://github.com/dmlc/mshadow.git mshadow
//Edit the Mshadow Makefile after line 122
MSHADOW_CFLAGS += -DMSHADOW_USE_PASCAL=1
$cd googletest
$git clone https://github.com/google/googletest.git
$cd ../../
$mkdir release; cd release
$cp ../make/crosscompile.jetson.mk ../make/config.mk
$make -j $(nproc)
Step3. Compile and install
由于缺失太多內(nèi)容颊埃，導(dǎo)致編譯一直報(bào)錯(cuò)
Bug1. kvstore_utils.h:27:26: fatal error: dmlc/logging.h: No such file or directory
Bug2.
