Keras.NET（1）- 安装

0.写在前面：

作者基于tensorflow 2.X，利用自在的keras接口完成了神经网络的训练，现在想把模型跑在c#端，但使用TensorFlowsharp好像还是tensorflow 1.X的语言风格，改动起来较为困难。所以想寻找c#接口的keras。经过一番搜索，找到两个资源，分别为keras-sharp（https://github.com/tcmxx/keras-sharp）和Keras.NET（https://github.com/SciSharp/Keras.NET/），keras-sharp看起来还久没有维护和更新了，所以放弃，尝试使用Keras.NET。

1.依赖项的安装：

• Python 2.7 - 3.7, Link: https://www.python.org/downloads/
• Install keras, numpy and one of the backend (Tensorflow/CNTK/Theano). Please see on how to configure: https://keras.io/backend/

 我是安装的python3.7.0（别忘了点击添加到环境变量），

python3.7.0安装完毕之后使用pip命令依次安装numpy，tensorflow，keras即可。这里注意：安装之后最好升级一下pip，

python -m pip install --upgrade pip
           

然后我是安装的tensorflow2.1 cpu版本，使用:

pip install tensorflow-cpu==2.1.0 -i https://mirrors.aliyun.com/pypi/simple
           

2.Keras.NET的安装

2.1 通过Nuget命令行安装

Install-Package Keras.NET
           

2.2 通过Nuget界面安装

安装完成之后自动添加了Keras，Numpy.Bare，Python.Runtime等

3.测试

拿着官方的minist例程测试一下，代码如下：

using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using System.Windows.Forms;

using Keras.Datasets;
using System;
using System.Collections.Generic;
using System.Text;
using Numpy;
using K = Keras.Backend;
using Keras;
using Keras.Models;
using Keras.Layers;
using Keras.Utils;
using Keras.Optimizers;
using System.IO;



namespace keras.net
{
    public partial class Form1 : Form
    {
        public Form1()
        {
            InitializeComponent();
        }

        private void Form1_Load(object sender, EventArgs e)
        {

        }

        private void test()
        {

            int batch_size = 200;
            int num_classes = 10;
            int epochs = 10;

            // input image dimensions
            int img_rows = 28, img_cols = 28;

            Shape input_shape = null;

            // the data, split between train and test sets
            var ((x_train, y_train), (x_test, y_test)) = MNIST.LoadData();

            if (K.ImageDataFormat() == "channels_first")
            {
                x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols);
                x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols);
                input_shape = (1, img_rows, img_cols);
            }
            else
            {
                x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1);
                x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1);
                input_shape = (img_rows, img_cols, 1);
            }

            x_train = x_train.astype(np.float32);
            x_test = x_test.astype(np.float32);
            x_train /= 255;
            x_test /= 255;
            Console.WriteLine("x_train shape: " + x_train.shape);
            Console.WriteLine(x_train.shape[0] + " train samples");
            Console.WriteLine(x_test.shape[0] + " test samples");

            // convert class vectors to binary class matrices
            y_train = Util.ToCategorical(y_train, num_classes);
            y_test = Util.ToCategorical(y_test, num_classes);

            // Build CNN model
            var model = new Sequential();
            model.Add(new Conv2D(32, kernel_size: (3, 3).ToTuple(),
                                 activation: "relu",
                                 input_shape: input_shape));
            model.Add(new Conv2D(64, (3, 3).ToTuple(), activation: "relu"));
            model.Add(new MaxPooling2D(pool_size: (2, 2).ToTuple()));
            model.Add(new Dropout(0.25));
            model.Add(new Flatten());
            model.Add(new Dense(128, activation: "relu"));
            model.Add(new Dropout(0.5));
            model.Add(new Dense(num_classes, activation: "softmax"));

            model.Compile(loss: "categorical_crossentropy",
              optimizer: new Adadelta(), metrics: new string[] { "accuracy" });

            model.Fit(x_train, y_train,
                      batch_size: batch_size,
                      epochs: epochs,
                      verbose: 1,
                      validation_data: new NDarray[] { x_test, y_test });

            model.Save("model.h5");
            model.SaveTensorflowJSFormat("./");

            var score = model.Evaluate(x_test, y_test, verbose: 0);
            Console.WriteLine("Test loss:" + score[0]);
            Console.WriteLine("Test accuracy:" + score[1]);
        }

        private void button1_Click(object sender, EventArgs e)
        {
            test();
        }
    }
}
           

程序会下载mnist数据集，非常慢又经常失败，从控制台的输出信息可以找到下载原网址如下：

https://s3.amazonaws.com/img-datasets/mnist.npz

我们可以随便从网上找一个用于keras的mnist数据集下载即可，我是用的这个https://download.csdn.net/download/lsldd/11502401

下载完成之后将mnist.npz文件放到keras数据集路径下（我的电脑是C:\Users\Administrator\.keras\datasets）

10个epoch之后损失和准确率如下：

注意：出现错误时，可以先试试解决方案平台改成X64，还不行的话删除所有依赖（删干净！）重新安装，不要使用anaconda安装依赖包！

ref：

https://blog.csdn.net/kinfey/article/details/99691593

https://scisharp.github.io/Keras.NET/index.html

https://github.com/SciSharp/Numpy.NET