Windows10上编译MXNet源码操作步骤(Python)

迈不过友情╰ 2022-04-04 12:58 464阅读 0赞

1. 按照[https://blog.csdn.net/fengbingchun/article/details/84997490][https_blog.csdn.net_fengbingchun_article_details_84997490]  中操作步骤首先在Windows上通过VS2017编译MXNet源代码；

2. 从  [https://mxnet.incubator.apache.org/install/windows\_setup.html\#install-the-mxnet-package-for-python][https_mxnet.incubator.apache.org_install_windows_setup.html_install-the-mxnet-package-for-python] 下载mingw64\_dll.zip；

3. 解压缩mingw64\_all.zip，并将其内的libgcc\_s\_seh-1.dll,libgfortran-dll,libquadmath-0.dll三个动态库拷贝到E:\\GitCode\\MXNet\_Test\\lib\\rel\\x64目录下，此目录下已包含由MXNet\_Test.sln工程编译生成的libopenblas.dll和libmxnet.dll动态库；

4. 将E:\\GitCode\\MXNet\_Test\\lib\\rel\\x64目录添加到系统环境变量，重启；

5. Windows中Python3.x, NumPy, Matplotlib的安装可参考：[https://blog.csdn.net/fengbingchun/article/details/77887050][https_blog.csdn.net_fengbingchun_article_details_77887050] ；

6. Windows中OpenCV3.4.2+Python3.2的安装可参考：[https://blog.csdn.net/fengbingchun/article/details/84030309][https_blog.csdn.net_fengbingchun_article_details_84030309] ；

7. 将E:\\GitCode\\MXNet\_Test\\lib\\rel\\x64目录下的libmxnet.dll动态库拷贝到E:\\GitCode\\MXNet\_Test\\src\\mxnet\\python\\mxnet目录下；

8. 打开命令提示符，并将其定位到E:\\GitCode\\MXNet\_Test\\src\\mxnet\\python目录下，执行：

python setup.py install

执行结果如下：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2ZlbmdiaW5nY2h1bg_size_16_color_FFFFFF_t_70][]

9. 验证MXNet python是否安装正确，测试代码如下：

import mxnet as mx
    import numpy as np
    import cv2
    
    img = cv2.imread("E:\GitCode\MXNet_Test\data\images\handwritten_digits\0.png")
    a = mx.nd.ones((2, 3)); print((a*2).asnumpy())
    b = np.array([1, 2, 3]); print(b)
    
    print("ok")

执行结果如下：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2ZlbmdiaW5nY2h1bg_size_16_color_FFFFFF_t_70 1][]

**GitHub**：[https://github.com/fengbingchun/MXNet\_Test][https_github.com_fengbingchun_MXNet_Test]

[https_blog.csdn.net_fengbingchun_article_details_84997490]: https://blog.csdn.net/fengbingchun/article/details/84997490
[https_mxnet.incubator.apache.org_install_windows_setup.html_install-the-mxnet-package-for-python]: https://mxnet.incubator.apache.org/install/windows_setup.html#install-the-mxnet-package-for-python
[https_blog.csdn.net_fengbingchun_article_details_77887050]: https://blog.csdn.net/fengbingchun/article/details/77887050
[https_blog.csdn.net_fengbingchun_article_details_84030309]: https://blog.csdn.net/fengbingchun/article/details/84030309
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2ZlbmdiaW5nY2h1bg_size_16_color_FFFFFF_t_70]: /images/20220404/46b30d1231f346e6acc9ee52b7cc0654.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2ZlbmdiaW5nY2h1bg_size_16_color_FFFFFF_t_70 1]: /images/20220404/8dc9ad389f434ab2bf593be7f3a8ef29.png
[https_github.com_fengbingchun_MXNet_Test]: https://github.com/fengbingchun/MXNet_Test