SciSharp/Keras.NETPublic

NotificationsYou must be signed in to change notification settings
Fork130
Star621

Keras.NET is a high-level neural networks API for C# and F#, with Python Binding and capable of running on top of TensorFlow, CNTK, or Theano.

License

MIT license

621 stars 130 forks Branches Tags Activity

Star

Notifications

You must be signed in to change notification settings

Branches Tags

Folders and files

Name		Name	Last commit message	Last commit date
Latest commit History 161 Commits
.github/workflows		.github/workflows
Examples		Examples
Images		Images
Keras.UnitTest		Keras.UnitTest
Keras		Keras
ReleaseBot		ReleaseBot
Tensorflow		Tensorflow
doc-gen		doc-gen
docs		docs
.gitignore		.gitignore
Keras.NET.sln		Keras.NET.sln
LICENSE		LICENSE
README.md		README.md
codemeta.json		codemeta.json

Repository files navigation

Keras.NET is a high-level neural networks API for C# and F# via a Python binding and capable of running on top of TensorFlow, CNTK, or Theano. It was developed with a focus on enabling fast experimentation. Being able to go from idea to result with the least possible delay is key to doing good research.

Use Keras if you need a deep learning library that:

Allows for easy and fast prototyping (through user friendliness, modularity, and extensibility).
Supports both convolutional networks and recurrent networks, as well as combinations of the two.
Runs seamlessly on CPU and GPU.

Keras.NET is using:

Prerequisite

Python 3.7 or 3.8, Link:https://www.python.org/downloads/
Install keras,numpy and one of the backends (Tensorflow/CNTK/Theano). Keras is now bundled with Tensorflow 2.0, so the easiest way to install Keras and Tensorflow at the same time is to simply installTensorflow 2.0.

Nuget

Install from nuget:https://www.nuget.org/packages/Keras.NET

dotnet add package Keras.NET

Example with XOR sample (C#)

//Load train dataNDarrayx=np.array(newfloat[,]{{0,0},{0,1},{1,0},{1,1}});NDarrayy=np.array(newfloat[]{0,1,1,0});//Build sequential modelvarmodel=newSequential();model.Add(newDense(32,activation:"relu",input_shape:newShape(2)));model.Add(newDense(64,activation:"relu"));model.Add(newDense(1,activation:"sigmoid"));//Compile and trainmodel.Compile(optimizer:"sgd",loss:"binary_crossentropy",metrics:newstring[]{"accuracy"});model.Fit(x,y,batch_size:2,epochs:1000,verbose:1);//Save model and weightsstringjson=model.ToJson();File.WriteAllText("model.json",json);model.SaveWeight("model.h5");//Load model and weightvarloaded_model=Sequential.ModelFromJson(File.ReadAllText("model.json"));loaded_model.LoadWeight("model.h5");

Output:

MNIST CNN Example (C#)

Python example taken from:https://keras.io/examples/mnist_cnn/

intbatch_size=128;intnum_classes=10;intepochs=12;// input image dimensionsintimg_rows=28,img_cols=28;Shapeinput_shape=null;// the data, split between train and test setsvar((x_train,y_train),(x_test,y_test))=MNIST.LoadData();if(Backend.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($"{x_test.shape[0]} test samples");// convert class vectors to binary class matricesy_train=Util.ToCategorical(y_train,num_classes);y_test=Util.ToCategorical(y_test,num_classes);// Build CNN modelvarmodel=newSequential();model.Add(newConv2D(32,kernel_size:(3,3).ToTuple(),activation:"relu",input_shape:input_shape));model.Add(newConv2D(64,(3,3).ToTuple(),activation:"relu"));model.Add(newMaxPooling2D(pool_size:(2,2).ToTuple()));model.Add(newDropout(0.25));model.Add(newFlatten());model.Add(newDense(128,activation:"relu"));model.Add(newDropout(0.5));model.Add(newDense(num_classes,activation:"softmax"));model.Compile(loss:"categorical_crossentropy",optimizer:newAdadelta(),metrics:newstring[]{"accuracy"});model.Fit(x_train,y_train,batch_size:batch_size,epochs:epochs,verbose:1,validation_data:newNDarray[]{x_test,y_test});varscore=model.Evaluate(x_test,y_test,verbose:0);Console.WriteLine($"Test loss:{score[0]}");Console.WriteLine($"Test accuracy:{score[1]}");