README.md

   1 # NNtrainer
   2
   3 [![Code Coverage](http://ci.nnstreamer.ai/nntrainer/ci/badge/codecoverage.svg)](http://ci.nnstreamer.ai/nntrainer/ci/gcov_html/index.html)
   4 ![GitHub repo size](https://img.shields.io/github/repo-size/nnstreamer/nntrainer)
   5 ![GitHub issues](https://img.shields.io/github/issues/nnstreamer/nntrainer)
   6 ![GitHub pull requests](https://img.shields.io/github/issues-pr/nnstreamer/nntrainer)
   7 <a href="https://scan.coverity.com/projects/nnstreamer-nntrainer">
   8   <img alt="Coverity Scan Build Status"
   9        src="https://scan.coverity.com/projects/22512/badge.svg"/>
  10 </a>
  11 [![DailyBuild](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/badge/daily_build_test_result_badge.svg)](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/build_result/)
  12
  13 NNtrainer is a Software Framework for training Neural Network models on devices.
  14
  15 ## Overview
  16
  17 NNtrainer is an Open Source Project. The aim of the NNtrainer is to develop a Software Framework to train neural network models on embedded devices which have relatively limited resources. Rather than training whole layers of a network from the scratch, NNtrainer finetunes the neural network model on device with user data for the personalization.
  18
  19 Even if NNtariner runs on device, it provides full functionalities to train models and also utilizes limited device resources efficiently. NNTrainer is able to train various machine learning algorithms such as k-Nearest Neighbor (k-NN), Neural Networks, Logistic Regression, Reinforcement Learning algorithms, Recurrent network and more. We also provide examples for various tasks such as Few-shot learning, ResNet, VGG, Product Rating and more will be added. All of these were tested on Samsung Galaxy smart phone with Android and PC (Ubuntu 18.04/20.04).
  20
  21 [ NNTrainer: Light-Weight On-Device Training Framework ](https://arxiv.org/pdf/2206.04688.pdf), arXiv, 2022 <br />
  22 [ NNTrainer: Towards the on-device learning for personalization ](https://www.youtube.com/watch?v=HWiV7WbIM3E), Samsung Software Developer Conference 2021 (Korean) <br />
  23 [ NNTrainer: Personalize neural networks on devices! ](https://www.youtube.com/watch?v=HKKowY78P1A), Samsung Developer Conference 2021 <br />
  24 [ NNTrainer: "On-device learning" ](https://www.youtube.com/embed/Jy_auavraKg?start=4035&end=4080), Samsung AI Forum 2021
  25
  26 ## Official Releases
  27
  28 |     | [Tizen](http://download.tizen.org/snapshots/tizen/unified/latest/repos/standard/packages/) | [Ubuntu](https://launchpad.net/~nnstreamer/+archive/ubuntu/ppa) | Android/NDK Build |
  29 | :-- | :--: | :--: | :--: |
  30 |     | 6.0M2 and later | 18.04 | 9/P |
  31 | arm | [![armv7l badge](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/badge/tizen.armv7l_result_badge.svg)](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/build_result/) | Available  | Ready |
  32 | arm64 |  [![aarch64 badge](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/badge/tizen.aarch64_result_badge.svg)](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/build_result/) | Available  | [![android badge](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/badge/arm64_v8a_android_result_badge.svg)](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/build_result/) |
  33 | x64 | [![x64 badge](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/badge/tizen.x86_64_result_badge.svg)](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/build_result/)  | [![ubuntu badge](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/badge/ubuntu_result_badge.svg)](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/build_result/)  | Ready  |
  34 | x86 | [![x86 badge](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/badge/tizen.i586_result_badge.svg)](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/build_result/)  | N/A  | N/A  |
  35 | Publish | [Tizen Repo](http://download.tizen.org/snapshots/tizen/unified/latest/repos/standard/packages/) | [PPA](https://launchpad.net/~nnstreamer/+archive/ubuntu/ppa) |   |
  36 | API | C (Official) | C/C++ | C/C++  |
  37
  38 - Ready: CI system ensures build-ability and unit-testing. Users may easily build and execute. However, we do not have automated release & deployment system for this instance.
  39 - Available: binary packages are released and deployed automatically and periodically along with CI tests.
  40 - [Daily Release](http://ci.nnstreamer.ai/nntrainer/ci/daily-build/build_result/)
  41 - SDK Support: Tizen Studio (6.0 M2+)
  42
  43 ## Maintainer
  44 * [Jijoong Moon](https://github.com/jijoongmoon)
  45 * [MyungJoo Ham](https://github.com/myungjoo)
  46 * [Geunsik Lim](https://github.com/leemgs)
  47
  48 ## Reviewers
  49 * [Sangjung Woo](https://github.com/again4you)
  50 * [Wook Song](https://github.com/wooksong)
  51 * [Jaeyun Jung](https://github.com/jaeyun-jung)
  52 * [Hyoungjoo Ahn](https://github.com/helloahn)
  53 * [Parichay Kapoor](https://github.com/kparichay)
  54 * [Dongju Chae](https://github.com/dongju-chae)
  55 * [Gichan Jang](https://github.com/gichan-jang)
  56 * [Yongjoo Ahn](https://github.com/anyj0527)
  57 * [Jihoon Lee](https://github.com/zhoonit)
  58 * [Hyeonseok Lee](https://github.com/lhs8928)
  59 * [Mete Ozay](https://github.com/meteozay)
  60 * [Hyunil Park](https://github.com/songgot)
  61 * [Jiho Chu](https://github.com/jihochu)
  62 * [Yelin Jeong](https://github.com/niley7464)
  63 * [Donghak Park](https://github.com/DonghakPark)
  64 * [Hyungjun Seo](https://github.com/SeoHyungjun)
  65 * [Seungbaek Hong](https://github.com/baek2sm)
  66 * [Sungsik Kong](https://github.com/skykongkong8)
  67 * [Donghyeon Jeong](https://github.com/djeong20)
  68
  69
  70 ## Components
  71
  72 ### Supported Layers
  73
  74 This component defines layers which consist of a neural network model. Layers have their own properties to be set.
  75
  76  | Keyword | Layer Class Name | Description |
  77  |:-------:|:---:|:---|
  78  | conv1d | Conv1DLayer | Convolution 1-Dimentional Layer |
  79  | conv2d | Conv2DLayer |Convolution 2-Dimentional Layer |
  80  | pooling2d | Pooling2DLayer |Pooling 2-Dimentional Layer. Support average / max / global average / global max pooling |
  81  | flatten | FlattenLayer | Flatten layer |
  82  | fully_connected | FullyConnectedLayer | Fully connected layer |
  83  | input | InputLayer | Input Layer.  This is not always required. |
  84  | batch_normalization | BatchNormalizationLayer | Batch normalization layer |
  85  | layer_normalization | LayerNormalizationLayer | Layer normalization layer |
  86  | activation | ActivationLayer | Set by layer property |
  87  | addition | AdditionLayer | Add input input layers |
  88  | attention | AttentionLayer | Attenstion layer |
  89  | centroid_knn | CentroidKNN | Centroid K-nearest neighbor layer |
  90  | concat | ConcatLayer | Concatenate input layers |
  91  | multiout | MultiOutLayer | Multi-Output Layer |
  92  | backbone_nnstreamer | NNStreamerLayer | Encapsulate NNStreamer layer |
  93  | backbone_tflite | TfLiteLayer | Encapsulate tflite as a layer |
  94  | permute | PermuteLayer | Permute layer for transpose |
  95  | preprocess_flip | PreprocessFlipLayer | Preprocess random flip layer |
  96  | preprocess_l2norm | PreprocessL2NormLayer | Preprocess simple l2norm layer to normalize |
  97  | preprocess_translate | PreprocessTranslateLayer | Preprocess translate layer |
  98  | reshape | ReshapeLayer | Reshape tensor dimension layer |
  99  | split | SplitLayer | Split layer |
 100  | dropout | DropOutLayer | Dropout Layer |
 101  | embedding | EmbeddingLayer | Embedding Layer |
 102  | positional_encoding | PositionalEncodingLayer | Positional Encoding Layer |
 103  | rnn | RNNLayer | Recurrent Layer |
 104  | rnncell | RNNCellLayer | Recurrent Cell Layer |
 105  | gru | GRULayer | Gated Recurrent Unit Layer |
 106  | grucell | GRUCellLayer | Gated Recurrent Unit Cell Layer |
 107  | lstm | LSTMLayer | Long Short-Term Memory Layer |
 108  | lstmcell | LSTMCellLayer | Long Short-Term Memory Cell Layer |
 109  | zoneoutlstmcell | ZoneoutLSTMCellLayer | Zoneout Long Short-Term Memory Cell Layer |
 110  | time_dist | TimeDistLayer | Time distributed Layer |
 111  | multi_head_attention | MultiHeadAttentionLayer | Multi Head Attention Layer |
 112
 113
 114 ### Supported Optimizers
 115
 116 NNTrainer Provides
 117
 118  | Keyword | Optimizer Name | Description |
 119  |:-------:|:---:|:---:|
 120  | sgd | Stochastic Gradient Decent | - |
 121  | adam | Adaptive Moment Estimation | - |
 122
 123  | Keyword | Leanring Rate | Description |
 124  |:-------:|:---:|:---:|
 125  | exponential | exponential learning rate decay | - |
 126  | constant | constant learning rate | - |
 127  | step | step learning rate | - |
 128
 129 ### Supported Loss Functions
 130
 131 NNTrainer provides
 132
 133  | Keyword | Class Name | Description |
 134  |:-------:|:---:|:---:|
 135  | cross_sigmoid | CrossEntropySigmoidLossLayer | Cross entropy sigmoid loss layer |
 136  | cross_softmax | CrossEntropySoftmaxLossLayer | Cross entropy softmax loss layer |
 137  | constant_derivative | ConstantDerivativeLossLayer | Constant derivative loss layer |
 138  | mse | MSELossLayer | Mean square error loss layer |
 139  | kld | KLDLossLayer | Kullback-Leibler Divergence loss layer |
 140
 141 ### Supported Activation Functions
 142
 143 NNTrainer provides
 144
 145  | Keyword | Loss Name | Description |
 146  |:-------:|:---:|:---|
 147  | tanh | tanh function | set as layer property |
 148  | sigmoid | sigmoid function | set as layer property |
 149  | relu | relu function | set as layer property |
 150  | softmax | softmax function | set as layer property |
 151
 152 ### Tensor
 153
 154 Tensor is responsible for calculation of a layer. It executes several operations such as addition, division, multiplication, dot production, data averaging and so on. In order to accelerate  calculation speed, CBLAS (C-Basic Linear Algebra: CPU) and CUBLAS (CUDA: Basic Linear Algebra) for PC (Especially NVIDIA GPU) are implemented for some of the operations. Later, these calculations will be optimized.
 155 Currently, we support lazy calculation mode to reduce complexity for copying tensors during calculations.
 156
 157  | Keyword | Description |
 158  |:-------:|:---:|
 159  | 4D Tensor | B, C, H, W|
 160  | Add/sub/mul/div | - |
 161  | sum, average, argmax | - |
 162  | Dot, Transpose | - |
 163  | normalization, standardization | - |
 164  | save, read | - |
 165
 166 ### Others
 167
 168 NNTrainer provides
 169
 170  | Keyword | Loss Name | Description |
 171  |:-------:|:---:|:---|
 172  | weight_initializer | Weight Initialization | Xavier(Normal/Uniform), LeCun(Normal/Uniform),  HE(Normal/Uniform) |
 173  | weight_regularizer | weight decay ( L2Norm only ) | needs set weight_regularizer_param & type |
 174
 175 ### APIs
 176 Currently, we provide [C APIs](https://github.com/nnstreamer/nntrainer/blob/master/api/capi/include/nntrainer.h) for Tizen. [C++ APIs](https://github.com/nnstreamer/nntrainer/blob/master/api/ccapi/include) are also provided for other platform. Java & C# APIs will be provided soon.
 177
 178
 179 ## [Getting Started](https://github.com/nnstreamer/nntrainer/blob/main/docs/getting-started.md)
 180
 181 Instructions for installing NNTrainer.
 182
 183 ### [Running Examples](https://github.com/nnstreamer/nntrainer/blob/main/docs/how-to-run-examples.md)
 184
 185 Instructions for preparing NNTrainer for execution
 186
 187 ### [Examples for NNTrainer](https://github.com/nnstreamer/nntrainer/tree/main/Applications)
 188
 189 NNTrainer example for a variety of networks
 190
 191 ## Open Source License
 192
 193 The NNtrainer is an open source project released under the terms of the Apache License version 2.0.
 194
 195 ## Contributing
 196
 197 Contributions are welcome! Please see our [Contributing](https://github.com/nnstreamer/nntrainer/blob/main/docs/contributing.md) Guide for more details.
 198
 199 [![](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/images/0)](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/links/0)[![](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/images/1)](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/links/1)[![](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/images/2)](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/links/2)[![](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/images/3)](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/links/3)[![](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/images/4)](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/links/4)[![](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/images/5)](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/links/5)[![](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/images/6)](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/links/6)[![](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/images/7)](https://sourcerer.io/fame/dongju-chae/nnstreamer/nntrainer/links/7)