diff --git a/README.md b/README.md index 40ca1afca..66b00d62f 100644 --- a/README.md +++ b/README.md @@ -9,12 +9,14 @@ [![Badge](https://img.shields.io/badge/link-996.icu-red.svg)](https://996.icu/#/en_US) [![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/javiercp/BinderTF.NET/master?urlpath=lab) -*master branch is based on tensorflow v2.x, v0.6x branch is based on tensorflow v2.6, v0.15-tensorflow1.15 is from tensorflow1.15.* +English | [中文](docs/Readme-CN.md) + +*master branch is corresponding to tensorflow v2.10, v0.6x branch is from tensorflow v2.6, v0.15-tensorflow1.15 is from tensorflow1.15.* ![tensors_flowing](docs/assets/tensors_flowing.gif) -### Why TensorFlow in C# and F# ? +## Why Tensorflow.NET ? `SciSharp STACK`'s mission is to bring popular data science technology into the .NET world and to provide .NET developers with a powerful Machine Learning tool set without reinventing the wheel. Since the APIs are kept as similar as possible you can immediately adapt any existing TensorFlow code in C# or F# with a zero learning curve. Take a look at a comparison picture and see how comfortably a TensorFlow/Python script translates into a C# program with TensorFlow.NET. @@ -22,59 +24,60 @@ SciSharp's philosophy allows a large number of machine learning code written in Python to be quickly migrated to .NET, enabling .NET developers to use cutting edge machine learning models and access a vast number of TensorFlow resources which would not be possible without this project. -In comparison to other projects, like for instance [TensorFlowSharp](https://www.nuget.org/packages/TensorFlowSharp/) which only provide TensorFlow's low-level C++ API and can only run models that were built using Python, Tensorflow.NET also implements TensorFlow's high level API where all the magic happens. This computation graph building layer is still under active development. Once it is completely implemented you can build new Machine Learning models in C# or F#. +In comparison to other projects, like for instance [TensorFlowSharp](https://www.nuget.org/packages/TensorFlowSharp/) which only provide TensorFlow's low-level C++ API and can only run models that were built using Python, Tensorflow.NET makes it possible to build the pipeline of training and inference with pure C# and F#. Besides, Tensorflow.NET provides binding of Tensorflow.Keras to make it easy to transfer your code from python to .NET. + +[ML.NET](https://github.com/dotnet/machinelearning) also support using tensorflow as backend to train and infer your model, which provides better integration with .NET. + +## Documention -Go through the online docs [TensorFlow for .NET](https://scisharp.github.io/tensorflow-net-docs) before you get started with Machine Learning in .NET. +Introduction and simple examples：[Tensorflow.NET Documents](https://scisharp.github.io/tensorflow-net-docs) -### How to use +Detailed documention：[The Definitive Guide to Tensorflow.NET](https://tensorflownet.readthedocs.io/en/latest/FrontCover.html) -| TensorFlow | tf native1.14, cuda 10.0 | tf native 1.15, cuda 10.0 | tf native 2.3, cuda 10.1 | tf native 2.4, cuda 11 | -| -------------------------- | ------------- | -------------- | ------------- | ------------- | -| tf.net 0.4x, tf.keras 0.5 | | | | x | -| tf.net 0.3x, tf.keras 0.4 | | | x | | -| tf.net 0.2x | | x | x | | -| tf.net 0.15 | x | x | | | -| tf.net 0.14 | x | | | | +Examples：[TensorFlow.NET Examples](https://github.com/SciSharp/TensorFlow.NET-Examples) -Troubleshooting of running example or installation, please refer [here](tensorflowlib/README.md). +Troubleshooting of running example or installation：[Tensorflow.NET FAQ](tensorflowlib/README.md) -There are many examples reside at [TensorFlow.NET Examples](https://github.com/SciSharp/TensorFlow.NET-Examples) written in C# and F#. +## Usage -#### TensorFlow.net Version -` tf.net 0.4x -> tf native 2.4` -`tf.net 0.6x -> tf native 2.6` -`tf.net 0.7x -> tf native 2.7` -`...` +### Installation -#### C# Example +You can search the package name in NuGet Manager, or use the commands below in pckage manager console. + +The installation contains two parts, the first is the main body: -Install TF.NET and TensorFlow binary through NuGet. ```sh -### install tensorflow C#/F# binding +### Install Tensorflow.NET PM> Install-Package TensorFlow.NET -### install keras for tensorflow + +### Install Tensorflow.Keras PM> Install-Package TensorFlow.Keras +``` + +The second part is the computing support part. Only one of the following packages is needed, depending on your device and system. -### Install tensorflow binary -### For CPU version +``` +### Cpu version for Windows, Linux and Mac PM> Install-Package SciSharp.TensorFlow.Redist -### For GPU version (CUDA and cuDNN are required) +### Gpu version for Windows (CUDA and CUDNN are required) PM> Install-Package SciSharp.TensorFlow.Redist-Windows-GPU + +### Gpu version for Linux (CUDA and CUDNN are required) +PM> Install-Package SciSharp.TensorFlow.Redist-Linux-GPU ``` -Import TF.NET and Keras API in your project. + +Two simple examples are given here to introduce the basic usage of Tensorflow.NET. As you can see, it's easy to write C# code just like that in Python. + +### Example - Linear Regression in `Eager` mode ```csharp using static Tensorflow.Binding; using static Tensorflow.KerasApi; using Tensorflow; using Tensorflow.NumPy; -``` -Linear Regression in `Eager` mode: - -```csharp // Parameters var training_steps = 1000; var learning_rate = 0.01f; @@ -120,10 +123,15 @@ foreach (var step in range(1, training_steps + 1)) Run this example in [Jupyter Notebook](https://github.com/SciSharp/SciSharpCube). -Toy version of `ResNet` in `Keras` functional API: +### Example - Toy version of `ResNet` in `Keras` functional API ```csharp -var layers = new LayersApi(); +using static Tensorflow.Binding; +using static Tensorflow.KerasApi; +using Tensorflow; +using Tensorflow.NumPy; + +var layers = keras.layers; // input layer var inputs = keras.Input(shape: (32, 32, 3), name: "img"); // convolutional layer @@ -147,96 +155,67 @@ var model = keras.Model(inputs, outputs, name: "toy_resnet"); model.summary(); // compile keras model in tensorflow static graph model.compile(optimizer: keras.optimizers.RMSprop(1e-3f), - loss: keras.losses.CategoricalCrossentropy(from_logits: true), + loss: keras.losses.SparseCategoricalCrossentropy(from_logits: true), metrics: new[] { "acc" }); // prepare dataset var ((x_train, y_train), (x_test, y_test)) = keras.datasets.cifar10.load_data(); +// normalize the input x_train = x_train / 255.0f; -y_train = np_utils.to_categorical(y_train, 10); // training model.fit(x_train[new Slice(0, 2000)], y_train[new Slice(0, 2000)], - batch_size: 64, - epochs: 10, - validation_split: 0.2f); + batch_size: 64, + epochs: 10, + validation_split: 0.2f); +// save the model +model.save("./toy_resnet_model"); ``` -#### F# Example - -Linear Regression in `Eager` mode: - -```fsharp -#r "nuget: TensorFlow.Net" -#r "nuget: TensorFlow.Keras" -#r "nuget: SciSharp.TensorFlow.Redist" +The F# example for linear regression is available [here](docs/Example-fsharp.md). -open Tensorflow -open Tensorflow.NumPy -open type Tensorflow.Binding -open type Tensorflow.KerasApi +More adcanced examples could be found in [TensorFlow.NET Examples](https://github.com/SciSharp/TensorFlow.NET-Examples). -let tf = New() -tf.enable_eager_execution() +## Version Relationships -// Parameters -let training_steps = 1000 -let learning_rate = 0.01f -let display_step = 100 - -// Sample data -let train_X = - np.array(3.3f, 4.4f, 5.5f, 6.71f, 6.93f, 4.168f, 9.779f, 6.182f, 7.59f, 2.167f, - 7.042f, 10.791f, 5.313f, 7.997f, 5.654f, 9.27f, 3.1f) -let train_Y = - np.array(1.7f, 2.76f, 2.09f, 3.19f, 1.694f, 1.573f, 3.366f, 2.596f, 2.53f, 1.221f, - 2.827f, 3.465f, 1.65f, 2.904f, 2.42f, 2.94f, 1.3f) -let n_samples = train_X.shape.[0] - -// We can set a fixed init value in order to demo -let W = tf.Variable(-0.06f,name = "weight") -let b = tf.Variable(-0.73f, name = "bias") -let optimizer = keras.optimizers.SGD(learning_rate) - -// Run training for the given number of steps. -for step = 1 to (training_steps + 1) do - // Run the optimization to update W and b values. - // Wrap computation inside a GradientTape for automatic differentiation. - use g = tf.GradientTape() - // Linear regression (Wx + b). - let pred = W * train_X + b - // Mean square error. - let loss = tf.reduce_sum(tf.pow(pred - train_Y,2)) / (2 * n_samples) - // should stop recording - // compute gradients - let gradients = g.gradient(loss,struct (W,b)) +| TensorFlow.NET Versions | tensorflow 1.14, cuda 10.0 | tensorflow 1.15, cuda 10.0 | tensorflow 2.3, cuda 10.1 | tensorflow 2.4, cuda 11 | tensorflow 2.7, cuda 11 |tensorflow 2.10, cuda 11 | +| -------------------------- | ------------- | -------------- | ------------- | ------------- | ------------ | ------------ | +| tf.net 0.10x, tf.keras 0.10 | | | | | | x | +| tf.net 0.7x, tf.keras 0.7 | | | | | x | | +| tf.net 0.4x, tf.keras 0.5 | | | | x | | | +| tf.net 0.3x, tf.keras 0.4 | | | x | | | | +| tf.net 0.2x | | x | x | | | | +| tf.net 0.15 | x | x | | | | | +| tf.net 0.14 | x | | | | | | - // Update W and b following gradients. - optimizer.apply_gradients(zip(gradients, struct (W,b))) - if (step % display_step) = 0 then - let pred = W * train_X + b - let loss = tf.reduce_sum(tf.pow(pred-train_Y,2)) / (2 * n_samples) - printfn $"step: {step}, loss: {loss.numpy()}, W: {W.numpy()}, b: {b.numpy()}" +``` +tf.net 0.4x -> tf native 2.4 +tf.net 0.6x -> tf native 2.6 +tf.net 0.7x -> tf native 2.7 +tf.net 0.10x -> tf native 2.10 +... ``` -Read the book [The Definitive Guide to Tensorflow.NET](https://tensorflownet.readthedocs.io/en/latest/FrontCover.html) if you want to know more about TensorFlow for .NET under the hood. +## Contribution: -### Contribute: +Feel like contributing to one of the hottest projects in the Machine Learning field? Want to know how Tensorflow magically creates the computational graph? -Feel like contributing to one of the hottest projects in the Machine Learning field? Want to know how Tensorflow magically creates the computational graph? We appreciate every contribution however small. There are tasks for novices to experts alike, if everyone tackles only a small task the sum of contributions will be huge. +We appreciate every contribution however small! There are tasks for novices to experts alike, if everyone tackles only a small task the sum of contributions will be huge. You can: -* Let everyone know about this project -* Port Tensorflow unit tests from Python to C# or F# -* Port missing Tensorflow code from Python to C# or F# -* Port Tensorflow examples to C# or F# and raise issues if you come accross missing parts of the API -* Debug one of the unit tests that is marked as Ignored to get it to work -* Debug one of the not yet working examples and get it to work +- Star Tensorflow.NET or share it with others +- Tell us about the missing APIs compared to Tensorflow +- Port Tensorflow unit tests from Python to C# or F# +- Port Tensorflow examples to C# or F# and raise issues if you come accross missing parts of the API or BUG +- Debug one of the unit tests that is marked as Ignored to get it to work +- Debug one of the not yet working examples and get it to work +- Help us to complete the documentions. -### How to debug unit tests: + +#### How to debug unit tests: The best way to find out why a unit test is failing is to single step it in C# or F# and its corresponding Python at the same time to see where the flow of execution digresses or where variables exhibit different values. Good Python IDEs like PyCharm let you single step into the tensorflow library code. -### Git Knowhow for Contributors +#### Git Knowhow for Contributors Add SciSharp/TensorFlow.NET as upstream to your local repo ... ```git @@ -247,6 +226,7 @@ Please make sure you keep your fork up to date by regularly pulling from upstrea ```git git pull upstream master ``` + ### Support Buy our book to make open source project be sustainable [TensorFlow.NET实战](https://item.jd.com/13441549.html)

diff --git a/docs/Example-fsharp.md b/docs/Example-fsharp.md new file mode 100644 index 000000000..578543454 --- /dev/null +++ b/docs/Example-fsharp.md @@ -0,0 +1,55 @@ +Linear Regression in `Eager` mode: + +```fsharp +#r "nuget: TensorFlow.Net" +#r "nuget: TensorFlow.Keras" +#r "nuget: SciSharp.TensorFlow.Redist" + +open Tensorflow +open Tensorflow.NumPy +open type Tensorflow.Binding +open type Tensorflow.KerasApi + +let tf = New() +tf.enable_eager_execution() + +// Parameters +let training_steps = 1000 +let learning_rate = 0.01f +let display_step = 100 + +// Sample data +let train_X = + np.array(3.3f, 4.4f, 5.5f, 6.71f, 6.93f, 4.168f, 9.779f, 6.182f, 7.59f, 2.167f, + 7.042f, 10.791f, 5.313f, 7.997f, 5.654f, 9.27f, 3.1f) +let train_Y = + np.array(1.7f, 2.76f, 2.09f, 3.19f, 1.694f, 1.573f, 3.366f, 2.596f, 2.53f, 1.221f, + 2.827f, 3.465f, 1.65f, 2.904f, 2.42f, 2.94f, 1.3f) +let n_samples = train_X.shape.[0] + +// We can set a fixed init value in order to demo +let W = tf.Variable(-0.06f,name = "weight") +let b = tf.Variable(-0.73f, name = "bias") +let optimizer = keras.optimizers.SGD(learning_rate) + +// Run training for the given number of steps. +for step = 1 to (training_steps + 1) do + // Run the optimization to update W and b values. + // Wrap computation inside a GradientTape for automatic differentiation. + use g = tf.GradientTape() + // Linear regression (Wx + b). + let pred = W * train_X + b + // Mean square error. + let loss = tf.reduce_sum(tf.pow(pred - train_Y,2)) / (2 * n_samples) + // should stop recording + // compute gradients + let gradients = g.gradient(loss,struct (W,b)) + + // Update W and b following gradients. + optimizer.apply_gradients(zip(gradients, struct (W,b))) + + if (step % display_step) = 0 then + let pred = W * train_X + b + let loss = tf.reduce_sum(tf.pow(pred-train_Y,2)) / (2 * n_samples) + printfn $"step: {step}, loss: {loss.numpy()}, W: {W.numpy()}, b: {b.numpy()}" +``` \ No newline at end of file diff --git a/docs/README-CN.md b/docs/README-CN.md new file mode 100644 index 000000000..06be13d2a --- /dev/null +++ b/docs/README-CN.md @@ -0,0 +1,217 @@ +![logo](assets/tf.net.logo.png) + +**Tensorflow.NET**是AI框架[TensorFlow](https://www.tensorflow.org/)在.NET平台上的实现，支持C#和F#，可以用来搭建深度学习模型并进行训练和推理，并内置了Numpy API，可以用来进行其它科学计算。 + +Tensorflow.NET并非对于Python的简单封装，而是基于C API的pure C#实现，因此使用时无需额外的环境，可以很方便地用NuGet直接安装使用。并且dotnet团队提供的[ML.NET](https://github.com/dotnet/machinelearning)也依赖于Tensorflow.NET，支持调用Tensorflow.NET进行训练和推理，可以很方便地融入.NET生态。 + +与tensorflow相同，Tensorflow.NET也内置了Keras这一高级API，只要在安装Tensorflow.NET的同时安装Tensorflow.Keras就可以使用，Keras支持以模块化的方式调用模型，给模型的搭建提供了极大的便利。 + +[![Join the chat at https://gitter.im/publiclab/publiclab](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/sci-sharp/community) +[![Tensorflow.NET](https://ci.appveyor.com/api/projects/status/wx4td43v2d3f2xj6?svg=true)](https://ci.appveyor.com/project/Haiping-Chen/tensorflow-net) +[![NuGet](https://img.shields.io/nuget/dt/TensorFlow.NET.svg)](https://www.nuget.org/packages/TensorFlow.NET) +[![Documentation Status](https://readthedocs.org/projects/tensorflownet/badge/?version=latest)](https://tensorflownet.readthedocs.io/en/latest/?badge=latest) +[![Badge](https://img.shields.io/badge/link-996.icu-red.svg)](https://996.icu/#/en_US) +[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/javiercp/BinderTF.NET/master?urlpath=lab) + +中文 | [English](https://github.com/SciSharp/TensorFlow.NET#readme) + +*当前主分支与Tensorflow2.10版本相对应，支持Eager Mode，同时也支持v1的静态图。* + + +![tensors_flowing](assets/tensors_flowing.gif) + +## Why Tensorflow.NET? + +`SciSharp STACK`开源社区的目标是构建.NET平台下易用的科学计算库，而Tensorflow.NET就是其中最具代表性的仓库之一。在深度学习领域Python是主流，无论是初学者还是资深开发者，模型的搭建和训练都常常使用Python写就的AI框架，比如tensorflow。但在实际应用深度学习模型的时候，又可能希望用到.NET生态，亦或只是因为.NET是自己最熟悉的领域，这时候Tensorflow.NET就有显著的优点，因为它不仅可以和.NET生态很好地贴合，其API还使得开发者很容易将Python代码迁移过来。下面的对比就是很好的例子，Python代码和C#代码有着高度相似的API，这会使得迁移的时候无需做过多修改。 + +![pythn vs csharp](assets/syntax-comparision.png) + +除了高度相似的API外，Tensorflow.NET与tensorflow也已经打通数据通道，tensorflow训练并保存的模型可以在Tensorflow.NET中直接读取并继续训练或推理，反之Tensorflow.NET保存的模型也可以在tensorflow中读取，这大大方便了模型的训练和部署。 + +与其它类似的库比如[TensorFlowSharp](https://www.nuget.org/packages/TensorFlowSharp/)相比，Tensorflow.NET的实现更加完全，提供了更多的高级API，使用起来更为方便，更新也更加迅速。 + + +## 文档 + +基本介绍与简单用例：[Tensorflow.NET Documents](https://scisharp.github.io/tensorflow-net-docs) + +详细文档：[The Definitive Guide to Tensorflow.NET](https://tensorflownet.readthedocs.io/en/latest/FrontCover.html) + +例程：[TensorFlow.NET Examples](https://github.com/SciSharp/TensorFlow.NET-Examples) + +运行例程常见问题：[Tensorflow.NET FAQ](tensorflowlib/README.md) + +## 安装与使用 + +安装可以在NuGet包管理器中搜索包名安装，也可以用下面命令行的方式。 + +安装分为两个部分，第一部分是Tensorflow.NET的主体： + +```sh +### 安装Tensorflow.NET +PM> Install-Package TensorFlow.NET +### 安装Tensorflow.Keras +PM> Install-Package TensorFlow.Keras +``` + +第二部分是计算支持部分，只需要根据自己的设备和系统选择下面之一即可： + +``` +### CPU版本 +PM> Install-Package SciSharp.TensorFlow.Redist + +### Windows下的GPU版本（需要安装CUDA和CUDNN） +PM> Install-Package SciSharp.TensorFlow.Redist-Windows-GPU + +### Linux下的GPU版本（需要安装CUDA和CUDNN） +PM> Install-Package SciSharp.TensorFlow.Redist-Linux-GPU +``` + +下面给出两个简单的例子，更多例子可以在[TensorFlow.NET Examples]中查看。 + +### 简单例子（使用Eager Mode进行线性回归） + +```csharp +using static Tensorflow.Binding; +using static Tensorflow.KerasApi; +using Tensorflow; +using Tensorflow.NumPy; + +// Parameters +var training_steps = 1000; +var learning_rate = 0.01f; +var display_step = 100; + +// Sample data +var X = np.array(3.3f, 4.4f, 5.5f, 6.71f, 6.93f, 4.168f, 9.779f, 6.182f, 7.59f, 2.167f, + 7.042f, 10.791f, 5.313f, 7.997f, 5.654f, 9.27f, 3.1f); +var Y = np.array(1.7f, 2.76f, 2.09f, 3.19f, 1.694f, 1.573f, 3.366f, 2.596f, 2.53f, 1.221f, + 2.827f, 3.465f, 1.65f, 2.904f, 2.42f, 2.94f, 1.3f); +var n_samples = X.shape[0]; + +// We can set a fixed init value in order to demo +var W = tf.Variable(-0.06f, name: "weight"); +var b = tf.Variable(-0.73f, name: "bias"); +var optimizer = keras.optimizers.SGD(learning_rate); + +// Run training for the given number of steps. +foreach (var step in range(1, training_steps + 1)) +{ + // Run the optimization to update W and b values. + // Wrap computation inside a GradientTape for automatic differentiation. + using var g = tf.GradientTape(); + // Linear regression (Wx + b). + var pred = W * X + b; + // Mean square error. + var loss = tf.reduce_sum(tf.pow(pred - Y, 2)) / (2 * n_samples); + // should stop recording + // Compute gradients. + var gradients = g.gradient(loss, (W, b)); + + // Update W and b following gradients. + optimizer.apply_gradients(zip(gradients, (W, b))); + + if (step % display_step == 0) + { + pred = W * X + b; + loss = tf.reduce_sum(tf.pow(pred - Y, 2)) / (2 * n_samples); + print($"step: {step}, loss: {loss.numpy()}, W: {W.numpy()}, b: {b.numpy()}"); + } +} +``` + +这一用例也可以在[Jupyter Notebook Example](https://github.com/SciSharp/SciSharpCube)进行运行. + +### 简单例子（使用Keras搭建Resnet） + +```csharp +using static Tensorflow.Binding; +using static Tensorflow.KerasApi; +using Tensorflow; +using Tensorflow.NumPy; + +var layers = keras.layers; +// input layer +var inputs = keras.Input(shape: (32, 32, 3), name: "img"); +// convolutional layer +var x = layers.Conv2D(32, 3, activation: "relu").Apply(inputs); +x = layers.Conv2D(64, 3, activation: "relu").Apply(x); +var block_1_output = layers.MaxPooling2D(3).Apply(x); +x = layers.Conv2D(64, 3, activation: "relu", padding: "same").Apply(block_1_output); +x = layers.Conv2D(64, 3, activation: "relu", padding: "same").Apply(x); +var block_2_output = layers.Add().Apply(new Tensors(x, block_1_output)); +x = layers.Conv2D(64, 3, activation: "relu", padding: "same").Apply(block_2_output); +x = layers.Conv2D(64, 3, activation: "relu", padding: "same").Apply(x); +var block_3_output = layers.Add().Apply(new Tensors(x, block_2_output)); +x = layers.Conv2D(64, 3, activation: "relu").Apply(block_3_output); +x = layers.GlobalAveragePooling2D().Apply(x); +x = layers.Dense(256, activation: "relu").Apply(x); +x = layers.Dropout(0.5f).Apply(x); +// output layer +var outputs = layers.Dense(10).Apply(x); +// build keras model +var model = keras.Model(inputs, outputs, name: "toy_resnet"); +model.summary(); +// compile keras model in tensorflow static graph +model.compile(optimizer: keras.optimizers.RMSprop(1e-3f), + loss: keras.losses.SparseCategoricalCrossentropy(from_logits: true), + metrics: new[] { "acc" }); +// prepare dataset +var ((x_train, y_train), (x_test, y_test)) = keras.datasets.cifar10.load_data(); +// normalize the input +x_train = x_train / 255.0f; +// training +model.fit(x_train[new Slice(0, 2000)], y_train[new Slice(0, 2000)], + batch_size: 64, + epochs: 10, + validation_split: 0.2f); +// save the model +model.save("./toy_resnet_model"); +``` + +此外，Tensorflow.NET也支持用F#搭建上述模型进行训练和推理。 + +## Tensorflow.NET版本对应关系 + +| TensorFlow.NET Versions | tensorflow 1.14, cuda 10.0 | tensorflow 1.15, cuda 10.0 | tensorflow 2.3, cuda 10.1 | tensorflow 2.4, cuda 11 | tensorflow 2.10, cuda 11 | +| -------------------------- | ------------- | -------------- | ------------- | ------------- | ------------ | +| tf.net 0.7+, tf.keras 0.7+ | | | | | x | +| tf.net 0.4x, tf.keras 0.5 | | | | x | | +| tf.net 0.3x, tf.keras 0.4 | | | x | | | +| tf.net 0.2x | | x | x | | | +| tf.net 0.15 | x | x | | | | +| tf.net 0.14 | x | | | | | + +如果使用过程中发现有缺失的版本，请告知我们，谢谢！ + +请注意Tensorflow.NET与Tensorflow.Keras版本存在一一对应关系，请安装与Tensorflow.NET对应的Tensorflow.Keras版本。 + +## 参与我们的开发: + +我们欢迎任何人的任何形式的贡献！无论是文档中的错误纠正，新特性提议，还是BUG修复等等，都会使得Tensorflow.NET项目越来越好，Tensorflow.NET的全体开发者也会积极帮助解决您提出的问题。 + +下面任何一种形式都可以帮助Tensorflow.NET越来越好： + +* Star和分享Tensorflow.NET项目 +* 为Tensorflow.NET添加更多的用例 +* 在issue中告知我们Tensorflow.NET目前相比tensorflow缺少的API或者没有对齐的特性 +* 在issue中提出Tensorflow.NET存在的BUG或者可以改进的地方 +* 在待办事项清单中选择一个进行或者解决某个issue +* 帮助我们完善文档，这也十分重要 + + +## 支持我们 +我们推出了[TensorFlow.NET实战](https://item.jd.com/13441549.html)这本书，包含了Tensorflow.NET主要开发者编写的讲解与实战例程，欢迎您的购买，希望这本书可以给您带来帮助。 +

+ + + +

+ +## 联系我们 + +可以在 [Twitter](https://twitter.com/ScisharpStack), [Facebook](https://www.facebook.com/scisharp.stack.9), [Medium](https://medium.com/scisharp), [LinkedIn](https://www.linkedin.com/company/scisharp-stack/)中关注我们，也可以在[Gitter](https://gitter.im/sci-sharp/community)中与项目开发者以及其它使用者进行沟通交流，也欢迎在仓库中提起issue。 + +TensorFlow.NET is a part of [SciSharp STACK](https://scisharp.github.io/SciSharp/) +
+