PaddlePaddle provides a rich set of computational units to enable users to adopt a modular approach to solving various learning problems. In this repo, we demonstrate how to use PaddlePaddle to solve common machine learning tasks, providing several different neural network model that anyone can easily learn and use.
[AlexNet](./PaddleCV/image_classification/models)|图像分类经典模型|首次在CNN中成功的应用了ReLU、Dropout和LRN，并使用GPU进行运算加速|[ImageNet Classification with Deep Convolutional Neural Networks](https://www.researchgate.net/publication/267960550_ImageNet_Classification_with_Deep_Convolutional_Neural_Networks)
[VGG](./PaddleCV/image_classification/models)|图像分类经典模型|在AlexNet的基础上使用3*3小卷积核，增加网络深度，具有很好的泛化能力|[Very Deep ConvNets for Large-Scale Inage Recognition](https://arxiv.org/pdf/1409.1556.pdf)
| 模型名称 | 模型简介 | 数据集 | 评估指标 mAP |
[GoogleNet](./PaddleCV/image_classification/models)|图像分类经典模型|在不增加计算负载的前提下增加了网络的深度和宽度，性能更加优越|[Going deeper with convolutions](https://ieeexplore.ieee.org/document/7298594)
[Inception-v4](./PaddleCV/image_classification/models)|图像分类经典模型|更加deeper和wider的inception结构|[Inception-ResNet and the Impact of Residual Connections on Learning](http://arxiv.org/abs/1602.07261)
[MobileNet](./PaddleCV/image_classification/models)|轻量级网络模型|为移动和嵌入式设备提出的高效模型|[MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861)
[SSD](./PaddleCV/object_detection/README_cn.md)|单阶段目标检测器|在不同尺度的特征图上检测对应尺度的目标,可以方便地插入到任何一种标准卷积网络中|[SSD: Single Shot MultiBox Detector](https://arxiv.org/abs/1512.02325)
[YOLOv3](./PaddleCV/yolov3/README.md)|单阶段目标检测器|基于darknet53主干网络在多种尺度的特征图上进行端到端实时目标检测,检测速度快|[YOLOv3: An Incremental Improvement](https://arxiv.org/abs/1804.02767)
[Face Detector: PyramidBox](./PaddleCV/face_detection/README_cn.md)|基于SSD的单阶段人脸检测器|利用上下文信息解决困难人脸的检测问题，网络表达能力高，鲁棒性强|[PyramidBox: A Context-assisted Single Shot Face Detector](https://arxiv.org/pdf/1803.07737.pdf)
<br/>
[Faster RCNN](./PaddleCV/rcnn/README_cn.md)|典型的两阶段目标检测器|创造性地采用卷积网络自行产生建议框，并且和目标检测网络共享卷积网络，建议框数目减少，质量提高|[Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://arxiv.org/abs/1506.01497)
[ICNet](./PaddleCV/icnet)|图像实时语义分割模型|即考虑了速度，也考虑了准确性，在高分辨率图像的准确性和低复杂度网络的效率之间获得平衡|[ICNet for Real-Time Semantic Segmentation on High-Resolution Images](https://arxiv.org/abs/1704.08545)
[DCGAN](./PaddleCV/gan/c_gan)|图像生成模型|深度卷积生成对抗网络，将GAN和卷积网络结合起来，以解决GAN训练不稳定的问题|[Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://arxiv.org/pdf/1511.06434.pdf)
[TSN](./PaddleCV/video_classification)|视频分类模型|基于长范围时间结构建模，结合了稀疏时间采样策略和视频级监督来保证使用整段视频时学习得有效和高效|[Temporal Segment Networks: Towards Good Practices for Deep Action Recognition](https://arxiv.org/abs/1608.00859)
| [ICNet](https://github.com/PaddlePaddle/models/tree/develop/PaddleCV/icnet) | 主要用于图像实时语义分割，能够兼顾速度和准确性，易于线上部署 | Cityscape | Mean IoU=67.0% |
| [Simple Baselines](https://github.com/PaddlePaddle/models/tree/develop/PaddleCV/human_pose_estimation) | coco2018关键点检测项目亚军方案，网络结构非常简单，效果达到state of the art | COCO val2017 | AP = 72.7% |
[Transformer](./PaddleNLP/neural_machine_translation/transformer/README.md)|机器翻译模型|基于self-attention，计算复杂度小，并行度高，容易学习长程依赖，翻译效果更好|[Attention Is All You Need](https://arxiv.org/abs/1706.03762)
[BERT](https://github.com/PaddlePaddle/LARK/tree/develop/BERT)|语义表示模型|在多个 NLP 任务上取得 SOTA 效果，支持多卡多机训练，支持混合精度训练|[BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://arxiv.org/abs/1810.04805)
[ELMo](https://github.com/PaddlePaddle/LARK/tree/develop/ELMo)|语义表示模型|支持多卡训练，训练速度比主流实现快1倍，提供在中文词法分析任务上迁移学习的示例。|[ELMo: Embeddings from Language Models](https://arxiv.org/abs/1802.05365)
[LAC](https://github.com/baidu/lac/blob/master/README.md)|联合的词法分析模型|能够整体性地完成中文分词、词性标注、专名识别任务|[Chinese Lexical Analysis with Deep Bi-GRU-CRF Network](https://arxiv.org/abs/1807.01882)
[DAM](./PaddleNLP/dialogue_model_toolkit/deep_attention_matching)|语义匹配模型|百度自然语言处理部发表于ACL-2018的工作,用于检索式聊天机器人多轮对话中应答的选择|[Multi-Turn Response Selection for Chatbots with Deep Attention Matching Network](http://aclweb.org/anthology/P18-1103)
[LAC （**Lexical Analysis of Chinese**](https://github.com/PaddlePaddle/models/tree/develop/PaddleNLP/lexical_analysis)）百度自主研发中文特色模型词法分析任务，**输入是一个字符串，而输出是句子中的词边界和词性、实体类别。
[ERNIE （Embeddings from Language Models）](https://github.com/PaddlePaddle/LARK/tree/develop/ERNIE)百度自研的语义表示模型，通过建模海量数据中的词、实体及实体关系，学习真实世界的语义知识。相较于 BERT 学习原始语言信号，ERNIE直接对先验语义知识单元进行建模，增强了模型语义表示能力。
<br/>
**BERT**
[BERT（Bidirectional Encoder Representation from Transformers） ](https://github.com/PaddlePaddle/LARK/tree/develop/BERT)是一个迁移能力很强的通用语义表示模型， 以 Transformer 为网络基本组件，以双向 Masked Language Model和 Next Sentence Prediction 为训练目标，通过预训练得到通用语义表示，再结合简单的输出层，应用到下游的 NLP 任务，在多个任务上取得了 SOTA 的结果。
<br/>
**ELMo**
[ELMo(Embeddings from Language Models) ](https://github.com/PaddlePaddle/LARK/tree/develop/ELMo)是重要的通用语义表示模型之一，以双向 LSTM 为网路基本组件，以 Language Model 为训练目标，通过预训练得到通用的语义表示，将通用的语义表示作为 Feature 迁移到下游 NLP 任务中，会显著提升下游任务的模型性能。
[TagSpace](./PaddleRec/tagspace)|文本及标签的embedding表示学习模型|应用于工业级的标签推荐，具体应用场景有feed新闻标签推荐等|[#TagSpace: Semantic embeddings from hashtags](https://www.bibsonomy.org/bibtex/0ed4314916f8e7c90d066db45c293462)
[GRU4Rec](./PaddleRec/gru4rec)|个性化推荐模型|首次将RNN（GRU）运用于session-based推荐，相比传统的KNN和矩阵分解，效果有明显的提升|[Session-based Recommendations with Recurrent Neural Networks](https://arxiv.org/abs/1511.06939)
| 模型名称 | 模型简介 |
[SSR](./PaddleRec/ssr)|序列语义检索推荐模型|使用参考论文中的思想，使用多种时间粒度进行用户行为预测|[Multi-Rate Deep Learning for Temporal Recommendation](https://dl.acm.org/citation.cfm?id=2914726)
[DeepCTR](./PaddleRec/ctr/README.cn.md)|点击率预估模型|只实现了DeepFM论文中介绍的模型的DNN部分，DeepFM会在其他例子中给出|[DeepFM: A Factorization-Machine based Neural Network for CTR Prediction](https://arxiv.org/abs/1703.04247)
[Multiview-Simnet](./PaddleRec/multiview_simnet)|个性化推荐模型|基于多元视图，将用户和项目的多个功能视图合并为一个统一模型|[A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems](http://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/frp1159-songA.pdf)
[DQN](./PaddleRL/DeepQNetwork/README_cn.md)|深度Q网络|value based强化学习算法，第一个成功地将深度学习和强化学习结合起来的模型|[Human-level control through deep reinforcement learning](https://www.nature.com/articles/nature14236)
[DoubleDQN](./PaddleRL/DeepQNetwork/README_cn.md)|DQN的变体|将Double Q的想法应用在DQN上，解决过优化问题|[Font Size: Deep Reinforcement Learning with Double Q-Learning](https://www.aaai.org/ocs/index.php/AAAI/AAAI16/paper/viewPaper/12389)
[DuelingDQN](./PaddleRL/DeepQNetwork/README_cn.md)|DQN的变体|改进了DQN模型，提高了模型的性能|[Dueling Network Architectures for Deep Reinforcement Learning](http://proceedings.mlr.press/v48/wangf16.html)
<br/>
<br/>
## License
<br/>
This tutorial is contributed by [PaddlePaddle](https://github.com/PaddlePaddle/Paddle) and licensed under the [Apache-2.0 license](LICENSE).
