tensorflow-mnist配置运行环境

素颜马尾好姑娘i 2022-11-04 11:17 145阅读 0赞

为了学习tensorflow和深度学习，看到mnist是tensorflow的入门程序，所以先找到了[https://github.com/sugyan/tensorflow-mnist][https_github.com_sugyan_tensorflow-mnist]。

先看看README文件

$ cat README.md 
    # MNIST classification by TensorFlow #
    
    - [MNIST For ML Beginners](https://www.tensorflow.org/tutorials/mnist/beginners/)
    - [Deep MNIST for Experts](https://www.tensorflow.org/tutorials/mnist/pros/)
    
    ![screencast](https://cloud.githubusercontent.com/assets/80381/11339453/f04f885e-923c-11e5-8845-33c16978c54d.gif)
    
    ### Requirement ###
    
    - Python >=2.7 or >=3.4
      - TensorFlow >=1.0
    - Node >=6.9
    
    
    ### How to run ###
    
        $ pip install -r requirements.txt
        $ npm install
        $ gunicorn main:app --log-file=-
    
    
    ### Deploy to Heroku ###
    
        $ heroku apps:create [NAME]
        $ heroku buildpacks:add heroku/nodejs
        $ heroku buildpacks:add heroku/python
        $ git push heroku master
    
    or Heroku Button.

### 创建conda的用于mnist的环境 ###

$ conda create -n mnist python=2.7
    $ conda activate mnist
    $ pip install -r requirements.txt

会提示2.7不再维护了，不过也能用.

### npm install ###

$ npm install
    
    > tensorflow-mnist@1.0.0 postinstall /home/hui/data/tensorflow-mnist
    > gulp
    
    [15:18:35] Using gulpfile ~/data/tensorflow-mnist/gulpfile.js
    [15:18:35] Starting 'build'...
    [15:18:35] Finished 'build' after 232 ms
    [15:18:35] Starting 'default'...
    [15:18:35] Finished 'default' after 13 μs

### gunicorn main:app --log-file=- ###

[2021-02-08 15:19:18 +0000] [14780] [INFO] Starting gunicorn 19.6.0
    [2021-02-08 15:19:18 +0000] [14780] [INFO] Listening at: http://127.0.0.1:8000 (14780)
    [2021-02-08 15:19:18 +0000] [14780] [INFO] Using worker: sync
    [2021-02-08 15:19:18 +0000] [14788] [INFO] Booting worker with pid: 14788
    W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
    W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
    W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
    W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
    W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
    W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.

### 接下来干什么 ###

然后在浏览器里面输入http://127.0.0.1:8000/，就可以手写输入数字，如下图所示，输出regression和convolutional值。

![tensorflow-mnist][]

[https_github.com_sugyan_tensorflow-mnist]: https://github.com/sugyan/tensorflow-mnist
[tensorflow-mnist]: /images/20221023/ec270b669d1b4100acba42f0fc1089d5.png