Oneflow-Inc · nlqq · Aug 11, 2020 · Aug 12, 2020 · Aug 12, 2020 · Aug 12, 2020
diff --git a/Classification/cnns/README.md → ComputerVision/cnns/README.md b/Classification/cnns/README.md → ComputerVision/cnns/README.md
diff --git a/Classification/cnns/alexnet_model.py → ComputerVision/cnns/alexnet_model.py b/Classification/cnns/alexnet_model.py → ComputerVision/cnns/alexnet_model.py
diff --git a/Classification/cnns/config.py → ComputerVision/cnns/config.py b/Classification/cnns/config.py → ComputerVision/cnns/config.py
@@ -53,7 +53,7 @@ def str2bool(v):
     parser.add_argument('--node_ips', type=str_list, default=['192.168.1.13', '192.168.1.14'],
                         help='nodes ip list for training, devided by ",", length >= num_nodes')
 
-    parser.add_argument("--model", type=str, default="resnet50",
+    parser.add_argument("--model", type=str, default="resnext50",
                         help="resnet50")
     parser.add_argument(
         '--use_fp16',

diff --git a/...on/cnns/data/ILSVRC2012_val_00020287.JPEG → ...on/cnns/data/ILSVRC2012_val_00020287.JPEG b/...on/cnns/data/ILSVRC2012_val_00020287.JPEG → ...on/cnns/data/ILSVRC2012_val_00020287.JPEG
diff --git a/Classification/cnns/data/fish.jpg → ComputerVision/cnns/data/fish.jpg b/Classification/cnns/data/fish.jpg → ComputerVision/cnns/data/fish.jpg
diff --git a/Classification/cnns/data/tiger.jpg → ComputerVision/cnns/data/tiger.jpg b/Classification/cnns/data/tiger.jpg → ComputerVision/cnns/data/tiger.jpg
diff --git a/...cation/cnns/docs/resnet50_lr_schedule.png → ...Vision/cnns/docs/resnet50_lr_schedule.png b/...cation/cnns/docs/resnet50_lr_schedule.png → ...Vision/cnns/docs/resnet50_lr_schedule.png
diff --git a/...cnns/docs/resnet50_validation_acuracy.png → ...cnns/docs/resnet50_validation_acuracy.png b/...cnns/docs/resnet50_validation_acuracy.png → ...cnns/docs/resnet50_validation_acuracy.png
diff --git a/Classification/cnns/evaluate.sh → ComputerVision/cnns/evaluate.sh b/Classification/cnns/evaluate.sh → ComputerVision/cnns/evaluate.sh
diff --git a/...ion/cnns/imagenet1000_clsidx_to_labels.py → ...ion/cnns/imagenet1000_clsidx_to_labels.py b/...ion/cnns/imagenet1000_clsidx_to_labels.py → ...ion/cnns/imagenet1000_clsidx_to_labels.py
diff --git a/Classification/cnns/inception_model.py → ComputerVision/cnns/inception_model.py b/Classification/cnns/inception_model.py → ComputerVision/cnns/inception_model.py
diff --git a/Classification/cnns/inference.sh → ComputerVision/cnns/inference.sh b/Classification/cnns/inference.sh → ComputerVision/cnns/inference.sh
diff --git a/Classification/cnns/job_function_util.py → ComputerVision/cnns/job_function_util.py b/Classification/cnns/job_function_util.py → ComputerVision/cnns/job_function_util.py
diff --git a/Classification/cnns/mobilenet_v2_model.py → ComputerVision/cnns/mobilenet_v2_model.py b/Classification/cnns/mobilenet_v2_model.py → ComputerVision/cnns/mobilenet_v2_model.py
diff --git a/Classification/cnns/of_cnn_evaluate.py → ComputerVision/cnns/of_cnn_evaluate.py b/Classification/cnns/of_cnn_evaluate.py → ComputerVision/cnns/of_cnn_evaluate.py
diff --git a/Classification/cnns/of_cnn_inference.py → ComputerVision/cnns/of_cnn_inference.py b/Classification/cnns/of_cnn_inference.py → ComputerVision/cnns/of_cnn_inference.py
diff --git a/Classification/cnns/of_cnn_train_val.py → ComputerVision/cnns/of_cnn_train_val.py b/Classification/cnns/of_cnn_train_val.py → ComputerVision/cnns/of_cnn_train_val.py
diff --git a/Classification/cnns/ofrecord_util.py → ComputerVision/cnns/ofrecord_util.py b/Classification/cnns/ofrecord_util.py → ComputerVision/cnns/ofrecord_util.py
diff --git a/Classification/cnns/optimizer_util.py → ComputerVision/cnns/optimizer_util.py b/Classification/cnns/optimizer_util.py → ComputerVision/cnns/optimizer_util.py
diff --git a/Classification/cnns/resnet_model.py → ComputerVision/cnns/resnet_model.py b/Classification/cnns/resnet_model.py → ComputerVision/cnns/resnet_model.py
diff --git a/Classification/cnns/resnet_to_onnx.py → ComputerVision/cnns/resnet_to_onnx.py b/Classification/cnns/resnet_to_onnx.py → ComputerVision/cnns/resnet_to_onnx.py
diff --git a/Classification/cnns/resnext_model.py → ComputerVision/cnns/resnext_model.py b/Classification/cnns/resnext_model.py → ComputerVision/cnns/resnext_model.py
diff --git a/Classification/cnns/tools/README.md → ComputerVision/cnns/tools/README.md b/Classification/cnns/tools/README.md → ComputerVision/cnns/tools/README.md
diff --git a/...sification/cnns/tools/extract_trainval.sh → ...uterVision/cnns/tools/extract_trainval.sh b/...sification/cnns/tools/extract_trainval.sh → ...uterVision/cnns/tools/extract_trainval.sh
diff --git a/...magenet_2012_validation_synset_labels.txt → ...magenet_2012_validation_synset_labels.txt b/...magenet_2012_validation_synset_labels.txt → ...magenet_2012_validation_synset_labels.txt
diff --git a/...nns/tools/imagenet_lsvrc_2015_synsets.txt → ...nns/tools/imagenet_lsvrc_2015_synsets.txt b/...nns/tools/imagenet_lsvrc_2015_synsets.txt → ...nns/tools/imagenet_lsvrc_2015_synsets.txt
diff --git a/...fication/cnns/tools/imagenet_metadata.txt → ...erVision/cnns/tools/imagenet_metadata.txt b/...fication/cnns/tools/imagenet_metadata.txt → ...erVision/cnns/tools/imagenet_metadata.txt
diff --git a/...ification/cnns/tools/imagenet_ofrecord.py → ...terVision/cnns/tools/imagenet_ofrecord.py b/...ification/cnns/tools/imagenet_ofrecord.py → ...terVision/cnns/tools/imagenet_ofrecord.py
diff --git a/...ls/preprocess_imagenet_validation_data.py → ...ls/preprocess_imagenet_validation_data.py b/...ls/preprocess_imagenet_validation_data.py → ...ls/preprocess_imagenet_validation_data.py
diff --git a/...tion/cnns/tools/process_bounding_boxes.py → ...sion/cnns/tools/process_bounding_boxes.py b/...tion/cnns/tools/process_bounding_boxes.py → ...sion/cnns/tools/process_bounding_boxes.py
diff --git a/Classification/cnns/train.sh → ComputerVision/cnns/train.sh b/Classification/cnns/train.sh → ComputerVision/cnns/train.sh
diff --git a/Classification/cnns/util.py → ComputerVision/cnns/util.py b/Classification/cnns/util.py → ComputerVision/cnns/util.py
diff --git a/Classification/cnns/vgg_model.py → ComputerVision/cnns/vgg_model.py b/Classification/cnns/vgg_model.py → ComputerVision/cnns/vgg_model.py
diff --git a/LanguageModeling/BERT/README.md → NaturalLanguageProcessing/BERT/README.md b/LanguageModeling/BERT/README.md → NaturalLanguageProcessing/BERT/README.md
diff --git a/LanguageModeling/BERT/bert.py → NaturalLanguageProcessing/BERT/bert.py b/LanguageModeling/BERT/bert.py → NaturalLanguageProcessing/BERT/bert.py
diff --git a/LanguageModeling/BERT/classifier.py → NaturalLanguageProcessing/BERT/classifier.py b/LanguageModeling/BERT/classifier.py → NaturalLanguageProcessing/BERT/classifier.py
diff --git a/LanguageModeling/BERT/config.py → NaturalLanguageProcessing/BERT/config.py b/LanguageModeling/BERT/config.py → NaturalLanguageProcessing/BERT/config.py
diff --git a/...ageModeling/BERT/convert_tf_ckpt_to_of.py → ...eProcessing/BERT/convert_tf_ckpt_to_of.py b/...ageModeling/BERT/convert_tf_ckpt_to_of.py → ...eProcessing/BERT/convert_tf_ckpt_to_of.py
diff --git a/LanguageModeling/BERT/pretrain.py → NaturalLanguageProcessing/BERT/pretrain.py b/LanguageModeling/BERT/pretrain.py → NaturalLanguageProcessing/BERT/pretrain.py
diff --git a/LanguageModeling/BERT/run_classifier.py → ...LanguageProcessing/BERT/run_classifier.py b/LanguageModeling/BERT/run_classifier.py → ...LanguageProcessing/BERT/run_classifier.py
diff --git a/LanguageModeling/BERT/run_pretraining.py → ...anguageProcessing/BERT/run_pretraining.py b/LanguageModeling/BERT/run_pretraining.py → ...anguageProcessing/BERT/run_pretraining.py
diff --git a/LanguageModeling/BERT/run_squad.py → NaturalLanguageProcessing/BERT/run_squad.py b/LanguageModeling/BERT/run_squad.py → NaturalLanguageProcessing/BERT/run_squad.py
diff --git a/LanguageModeling/BERT/squad.py → NaturalLanguageProcessing/BERT/squad.py b/LanguageModeling/BERT/squad.py → NaturalLanguageProcessing/BERT/squad.py
diff --git a/LanguageModeling/BERT/squad_util.py → NaturalLanguageProcessing/BERT/squad_util.py b/LanguageModeling/BERT/squad_util.py → NaturalLanguageProcessing/BERT/squad_util.py
diff --git a/LanguageModeling/BERT/tokenization.py → ...alLanguageProcessing/BERT/tokenization.py b/LanguageModeling/BERT/tokenization.py → ...alLanguageProcessing/BERT/tokenization.py
diff --git a/LanguageModeling/BERT/util.py → NaturalLanguageProcessing/BERT/util.py b/LanguageModeling/BERT/util.py → NaturalLanguageProcessing/BERT/util.py
diff --git a/README.md b/README.md
@@ -1,24 +1,171 @@
 # OneFlow Deep Learning Benchmarks
-## Introduction
-This repository provides OneFlow deep learning benchmark examples for CV, CTR and NLP, and more models are on the way and will be provided here when ready.
-
-## [Convolutional Networks](./Classification/cnns) for Computer Vision Classification
-- [ResNet-50](./Classification/cnns)
-- [ResNeXt-50-32*4d](./Classification/cnns)
-- [VGG-16](./Classification/cnns)
-- [Inception-V3](./Classification/cnns)
-- [AlexNet](./Classification/cnns)
-- [MobileNet-V2](./Classification/cnns)
-
-## [Wide Deep Learning](./ClickThroughRate/WideDeepLearning) for Click-Through-Rate (CTR) Recommender Systems
-- [OneFlow-WDL](./ClickThroughRate/WideDeepLearning)
-
-## [BERT](./LanguageModeling/BERT) for Nature Language Process
-- [BERT Pretrain for Language Modeling](./LanguageModeling/BERT)
-- [SQuAD for Question Answering](./LanguageModeling/BERT)
-- [CoLA and MRPC of GLUE](./LanguageModeling/BERT)
-
-## OneFlow Benchmark Test Reports
+
+ [![](https://img.shields.io/badge/Language-CH-red.svg)](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/dev_sx_benchmark/README_CH.md)
+
+This repository provides a collection of example implementations and modeling solutions using the latest OneFlow's high-level APIs, for CV, CTR and NLP models as a benchmark.
+
+It aims to demonstrate the best practices for modeling so that OneFlow users can take full advantage of OneFlow for their research and product development.
+
+ More models are coming!
+
+## Contents
+
+### Models and Implementations
+
+- ### Computer Vision
+
+  - #### Image Classification
+
+| Model                                                        | Reference (Paper)                                            |
+| ------------------------------------------------------------ | ------------------------------------------------------------ |
+| [ResNet-50](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/resnet_model.py) | [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385) |
+| [ResNeXt](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/resnext_model.py) | [Aggregated_Residual_Transformations_CVPR_2017](https://openaccess.thecvf.com/content_cvpr_2017/papers/Xie_Aggregated_Residual_Transformations_CVPR_2017_paper.pdf) |
+| [VGG-16](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/vgg_model.py) | [VGG16 – Convolutional Network for Classification and Detection](https://neurohive.io/en/popular-networks/vgg16/) |
+| [Inception-V3](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/inception_model.py) | [Inception V3 Deep Convolutional Architecture For Classifying Acute Myeloid/Lymphoblastic Leukemia](https://software.intel.com/content/www/us/en/develop/articles/inception-v3-deep-convolutional-architecture-for-classifying-acute-myeloidlymphoblastic.html) |
+| [AlexNet](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/alexnet_model.py) | [ImageNet Classification with Deep Convolutional Neural Networks](http://vision.stanford.edu/teaching/cs231b_spring1415/slides/alexnet_tugce_kyunghee.pdf) |
+| [MobileNet-V2](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/mobilenet_v2_model.py) | [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861) |
+
+- ## Natural Language Processing
+
+| Model                                                        | Reference (Paper)                                            |
+| ------------------------------------------------------------ | ------------------------------------------------------------ |
+| [BERT (Bidirectional Encoder Representations from Transformers)](https://github.com/OneFlow/models/blob/master/official/nlp/bert) | [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://arxiv.org/abs/1810.04805) |
+| [SQuAD for Question Answering](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/LanguageModeling/BERT/run_squad.py) | [BERT-SQuAD](https://github.com/kamalkraj/BERT-SQuAD)        |
+| [CoLA and MRPC of GLUE](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/LanguageModeling/BERT/run_classifier.py) | [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://www.aclweb.org/anthology/W18-5446.pdf) |
+
+- ## Click-Through-Rate 
+
+| Model                                                        | Reference (Paper)                                            |
+| ------------------------------------------------------------ | ------------------------------------------------------------ |
+| [OneFlow-WDL](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/ClickThroughRate/WideDeepLearning) | [**Wide & Deep Learning for Recommender Systems**](https://arxiv.org/pdf/1606.07792.pdf) |
+
+
+
+## Get started with the models
+
+- The models in the master branch are developed using OneFlow [], and they target the OneFlow [nightly binaries](https://github.com/OneFlow/OneFlow#installation) built from the [master branch of OneFlow](https://github.com/OneFlow/OneFlow/tree/master).
+
+- The stable versions targeting releases of OneFlow are available as tagged branches or [downloadable releases](https://github.com/OneFlow/models/releases).
+
+
+
+Please follow the below steps before running models in this repository.
+
+### Requirements
+
+- Python >= 3.5
+
+- CUDA Toolkit Linux x86_64 Driver
+
+  | OneFlow       | CUDA Driver Version |
+  | ------------- | ------------------- |
+  | oneflow_cu102 | >= 440.33           |
+  | oneflow_cu101 | >= 418.39           |
+  | oneflow_cu100 | >= 410.48           |
+  | oneflow_cu92  | >= 396.26           |
+  | oneflow_cu91  | >= 390.46           |
+  | oneflow_cu90  | >= 384.81           |
+
+  - CUDA runtime is statically linked into OneFlow. OneFlow will work on a minimum supported driver, and any driver beyond. For more information, please refer to [CUDA compatibility documentation](https://docs.nvidia.com/deploy/cuda-compatibility/index.html).
+  - Support for latest stable version of CUDA will be prioritized. Please upgrade your Nvidia driver to version 440.33 or above and install `oneflow_cu102` if possible.
+  - We are sorry that due to limits on bandwidth and other resources, we could only guarantee the efficiency and stability of `oneflow_cu102`. We will improve it ASAP.
+
+### Installation
+
+#### Method 1: Install with pip package
+
+- To install latest release of OneFlow with CUDA support:
+
+  ```
+  python3 -m pip install --find-links https://oneflow-inc.github.io/nightly oneflow_cu102 --user
+  ```
+
+- To install OneFlow with legacy CUDA support, run one of:
+
+  ```
+  python3 -m pip install --find-links https://oneflow-inc.github.io/nightly oneflow_cu101 --user
+  python3 -m pip install --find-links https://oneflow-inc.github.io/nightly oneflow_cu100 --user
+  python3 -m pip install --find-links https://oneflow-inc.github.io/nightly oneflow_cu92 --user
+  python3 -m pip install --find-links https://oneflow-inc.github.io/nightly oneflow_cu91 --user
+  python3 -m pip install --find-links https://oneflow-inc.github.io/nightly oneflow_cu90 --user
+  ```
+
+- If you are in China, you could run this to have pip download packages from domestic mirror of pypi:
+
+  ```
+  python3 -m pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
+  ```
+
+  For more information on this, please refer to [pypi 镜像使用帮助](https://mirror.tuna.tsinghua.edu.cn/help/pypi/)
+
+- CPU-only OneFlow is not available for now.
+
+- Releases are built with G++/GCC 4.8.5, cuDNN 7 and MKL 2020.0-088.
+
+#### Method 2: Build from source
+
+1. System Requirements to Build OneFlow
+
+- Please use a newer version of CMake to build OneFlow. You could download cmake release from [here](https://github.com/Kitware/CMake/releases/download/v3.14.0/cmake-3.14.0-Linux-x86_64.tar.gz).
+
+- Please make sure you have G++ and GCC >= 4.8.5 installed. Clang is not supported for now.
+
+- To install dependencies, run:
+
+  ```
+  yum-config-manager --add-repo https://yum.repos.intel.com/setup/intelproducts.repo && \
+  rpm --import https://yum.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS-2019.PUB && \
+  yum update -y && yum install -y epel-release && \
+  yum install -y intel-mkl-64bit-2020.0-088 nasm swig rdma-core-devel
+  ```
+
+  On CentOS, if you have MKL installed, please update the environment variable:
+
+  ```
+  export LD_LIBRARY_PATH=/opt/intel/lib/intel64_lin:/opt/intel/mkl/lib/intel64:$LD_LIBRARY_PATH
+  ```
+
+  If you don't want to build OneFlow with MKL, you could install OpenBLAS:
+
+  ```
+  sudo yum -y install openblas-devel
+  ```
+
+2. Clone Source Code
+
+Clone source code and submodules (faster, recommended)
+
+```
+git clone https://github.com/Oneflow-Inc/oneflow
+cd oneflow
+git submodule update --init --recursive
+```
+
+Or you could also clone the repo with `--recursive` flag to clone third_party submodules together
+
+```
+git clone https://github.com/Oneflow-Inc/oneflow --recursive
+```
+
+3. Build and Install OneFlow
+
+```
+cd build
+cmake ..
+make -j$(nproc)
+make pip_install
+```
+
+- For pure CPU build, please add this CMake flag `-DBUILD_CUDA=OFF`.
+
+### More models to come
+
+[new models]
+
+## Contributions
+
+- How to add new models?
+- How to add new framework tests?
 
 | Model | DType | XLA | Throughput | Speedup on 32 devices |
 | ----- | ----- | --- | ---------- | ------- |

diff --git a/README_CH.md b/README_CH.md
@@ -0,0 +1,46 @@
+# OneFlow 深度学习基准
+
+ [![](https://img.shields.io/badge/Language-EN-blue.svg)](https://github.com/Oneflow-Inc/OneFlow-Benchmark/tree/dev_sx_benchmark)
+
+本仓库将提供一系列由 OneFlow 最新的高级接口实现的模型及样例网络，模型基准涉及计算机视觉（Computer Vision，CV）、点击率推荐（Click-Through-Rate，CTR）、自然语言处理（Natural Language Processing， NLP）。
+
+为了能让 OneFlow 用户实现自己拓展研究和产品迭代的需求，充分利用该框架，本仓库旨在提供各个模型基于 OneFlow 的最佳实现。
+
+更多新模型正在路上！
+
+## 内容
+
+### 模型及实现
+
+- ### 计算机视觉（Computer Vision）
+
+  - #### 图片识别（Image Classification）
+
+| 模型                                                         | 参考来源（论文）                                             |
+| ------------------------------------------------------------ | ------------------------------------------------------------ |
+| [ResNet-50](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/resnet_model.py) | [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385) |
+| [ResNeXt](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/resnext_model.py) | [Aggregated_Residual_Transformations_CVPR_2017](https://openaccess.thecvf.com/content_cvpr_2017/papers/Xie_Aggregated_Residual_Transformations_CVPR_2017_paper.pdf) |
+| [VGG-16](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/vgg_model.py) | [VGG16 – Convolutional Network for Classification and Detection](https://neurohive.io/en/popular-networks/vgg16/) |
+| [Inception-V3](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/inception_model.py) | [Inception V3 Deep Convolutional Architecture For Classifying Acute Myeloid/Lymphoblastic Leukemia](https://software.intel.com/content/www/us/en/develop/articles/inception-v3-deep-convolutional-architecture-for-classifying-acute-myeloidlymphoblastic.html) |
+| [AlexNet](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/alexnet_model.py) | [ImageNet Classification with Deep Convolutional Neural Networks](http://vision.stanford.edu/teaching/cs231b_spring1415/slides/alexnet_tugce_kyunghee.pdf) |
+| [MobileNet-V2](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/Classification/cnns/mobilenet_v2_model.py) | [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861) |
+
+- ### 自然语言处理（Natural Language Processing）
+
+| 模型                                                         | 参考来源（论文）                                             |
+| ------------------------------------------------------------ | ------------------------------------------------------------ |
+| [BERT (Bidirectional Encoder Representations from Transformers)](https://github.com/OneFlow/models/blob/master/official/nlp/bert) | [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://arxiv.org/abs/1810.04805) |
+| [SQuAD for Question Answering](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/LanguageModeling/BERT/run_squad.py) | [BERT-SQuAD](https://github.com/kamalkraj/BERT-SQuAD)        |
+| [CoLA and MRPC of GLUE](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/LanguageModeling/BERT/run_classifier.py) | [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://www.aclweb.org/anthology/W18-5446.pdf) |
+
+- ### 点击率推荐（Click-Through-Rate）
+
+  | 模型                                                         | 参考来源（论文）                                             |
+  | ------------------------------------------------------------ | ------------------------------------------------------------ |
+  | [OneFlow-WDL](https://github.com/Oneflow-Inc/OneFlow-Benchmark/blob/master/ClickThroughRate/WideDeepLearning) | [**Wide & Deep Learning for Recommender Systems**](https://arxiv.org/pdf/1606.07792.pdf) |
+
+
+
+### 开始使用模型
+
+...