diff --git a/models/contentunderstanding/classification/config.yaml b/models/contentunderstanding/classification/config.yaml
index 4cf4d1bb7dae64865e2a2738e80672ac23934702..d1748137f0c4d994b3a566debf43dbdc2c3d66dc 100644
--- a/models/contentunderstanding/classification/config.yaml
+++ b/models/contentunderstanding/classification/config.yaml
@@ -18,7 +18,7 @@ train:
     strategy: "async"
 
   epochs: 10
-  workspace: "paddlerec.models.contentunderstandin.classification"
+  workspace: "paddlerec.models.contentunderstanding.classification"
 
   reader:
     batch_size: 5
diff --git a/models/contentunderstanding/classification/model.py b/models/contentunderstanding/classification/model.py
index 6254199c0fd4ceec48ba7f7d8bee3382d967fd02..e4630820c868af8334fc8edfd2b6c1f4d9e77503 100644
--- a/models/contentunderstanding/classification/model.py
+++ b/models/contentunderstanding/classification/model.py
@@ -40,9 +40,12 @@ class Model(ModelBase):
         data = fluid.data(name="input", shape=[None, self.max_len], dtype='int64')
         label = fluid.data(name="label", shape=[None, 1], dtype='int64')
         seq_len = fluid.data(name="seq_len", shape=[None], dtype='int64')
+
+        self._data_var = [data, label, seq_len]
+
         # embedding layer
         emb = fluid.embedding(input=data, size=[self.dict_dim, self.emb_dim])
-        emb = fluid.layers.sequence_unpad(emb, length=self.seq_len)
+        emb = fluid.layers.sequence_unpad(emb, length=seq_len)
         # convolution layer
         conv = fluid.nets.sequence_conv_pool(
             input=emb,
@@ -52,7 +55,7 @@ class Model(ModelBase):
             pool_type="max")
 
         # full connect layer
-        fc_1 = fluid.layers.fc(input=[conv], size=hid_dim)
+        fc_1 = fluid.layers.fc(input=[conv], size=self.hid_dim)
         # softmax layer
         prediction = fluid.layers.fc(input=[fc_1], size=self.class_dim, act="softmax")
         cost = fluid.layers.cross_entropy(input=prediction, label=label)
@@ -60,18 +63,18 @@ class Model(ModelBase):
         acc = fluid.layers.accuracy(input=prediction, label=label) 
 
         self.cost = avg_cost
-        self.metrics["acc"] = cos_pos
+        self._metrics["acc"] = acc
 
     def get_cost_op(self):
         return self.cost
 
     def get_metrics(self):
-        return self.metrics
+        return self._metrics
 
     def optimizer(self):
         learning_rate = 0.01
         sgd_optimizer = fluid.optimizer.Adagrad(learning_rate=learning_rate)
         return sgd_optimizer
 
-    def infer_net(self, parameter_list):
+    def infer_net(self):
         self.train_net()
diff --git a/models/contentunderstanding/classification/reader.py b/models/contentunderstanding/classification/reader.py
index 9a93211ed6997c342c412b7c2a043f89838332da..f90097d702df461d226443c32570e46ea3a0b093 100644
--- a/models/contentunderstanding/classification/reader.py
+++ b/models/contentunderstanding/classification/reader.py
@@ -44,5 +44,9 @@ class TrainReader(Reader):
             if data is None:
                 yield None
                 return
+            data = [int(i) for i in data]
+            label = [int(i) for i in label]
+            seq_len = [int(i) for i in seq_len]
+            print >>sys.stderr, str([('data', data), ('label', label), ('seq_len', seq_len)])
             yield [('data', data), ('label', label), ('seq_len', seq_len)]
         return data_iter
diff --git a/models/contentunderstanding/readme.md b/models/contentunderstanding/readme.md
index 417fdde3a8c31a184db7d7dd0f372f769c9615c1..06be7106b287149d24b56773689dad08708a064f 100644
--- a/models/contentunderstanding/readme.md
+++ b/models/contentunderstanding/readme.md
@@ -71,13 +71,13 @@ python text2paddle.py raw_big_train_data/ raw_big_test_data/ train_big_data test
 ### 训练
 
 ```
-python -m paddlerec.run -m paddlerec.models.rank.dnn -d cpu -e single
+python -m paddlerec.run -m paddlerec.models.contentunderstanding.classification -d cpu -e single
 ```
 
 ### 预测
 
 ```
-python -m paddlerec.run -m paddlerec.models.rank.dnn -d cpu -e single
+python -m paddlerec.run -m paddlerec.models.contentunderstanding.classification -d cpu -e single
 ```
 
 ## 效果对比
diff --git a/models/contentunderstanding/tagspace/config.yaml b/models/contentunderstanding/tagspace/config.yaml
index e0232f3faf221be04a6f95a6b0d011d612d0e35c..70333fcbf7edf4b6b5f54145e29cb122ed3ae9c6 100644
--- a/models/contentunderstanding/tagspace/config.yaml
+++ b/models/contentunderstanding/tagspace/config.yaml
@@ -18,7 +18,7 @@ train:
     strategy: "async"
 
   epochs: 10
-  workspace: "paddlerec.models.rank.tagspace"
+  workspace: "paddlerec.models.contentunderstanding.tagspace"
 
   reader:
     batch_size: 5
diff --git a/readme.md b/readme.md
index 3dfbf8d3904d47092052b5d358dd5aeb4ce67b8a..f90176c5de9e90207209ed30a8b9cc53e4a50c4c 100644
--- a/readme.md
+++ b/readme.md
@@ -1,2 +1,171 @@
-# PaddleRec
-推荐算法，大规模并行训练支持
+<p align="center">
+<img align="center" src="doc/imgs/logo.png">
+<p>
+
+<p align="center">
+    <br>
+    <img alt="Release" src="https://img.shields.io/badge/Release-0.1.0-yellowgreen">
+    <img alt="License" src="https://img.shields.io/github/license/PaddlePaddle/Serving">
+    <img alt="Slack" src="https://img.shields.io/badge/Join-Slack-green">
+    <br>
+<p>
+
+
+<h2 align="center">什么是PaddleRec</h2>
+
+<p align="center">
+<img align="center" src="doc/imgs/structure.png">
+<p>
+
+- 源于飞桨生态的搜索推荐模型**一站式开箱即用工具** 
+- 适合初学者，开发者，研究者从调研，训练到预测部署的全流程解决方案
+- 包含语义理解、召回、粗排、精排、多任务学习、融合等多个任务的推荐搜索算法库
+- 配置**yaml**自定义选项，即可快速上手使用单机训练、大规模分布式训练、离线预测、在线部署
+
+
+<h2 align="center">PadlleRec概览</h2>
+
+<p align="center">
+<img align="center" src="doc/imgs/overview.png">
+<p>
+
+
+<h2 align="center">推荐系统-流程概览</h2>
+
+<p align="center">
+<img align="center" src="doc/imgs/rec-overview.png">
+<p>
+
+<h2 align="center">便捷安装</h2>
+
+### 环境要求
+* Python 2.7/ 3.5 / 3.6 / 3.7
+* PaddlePaddle  >= 1.7.2
+* 操作系统: Windows/Mac/Linux
+  
+### 安装命令
+
+- 安装方法一<PIP源直接安装>：
+  ```bash
+  python -m pip install paddle-rec
+  ```
+
+- 安装方法二
+
+  源码编译安装
+  1. 安装飞桨  **注：需要用户安装版本 >1.7.2 的飞桨**
+
+    ```shell
+    python -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
+    ```
+
+  2. 源码安装PaddleRec
+
+    ```
+    git clone https://github.com/PaddlePaddle/PaddleRec/
+    cd PaddleRec
+    python setup.py install
+    ```
+
+
+<h2 align="center">快速启动</h2>
+
+### 启动内置模型的默认配置
+
+目前框架内置了多个模型，简单的命令即可使用内置模型开始单机训练和本地1*1模拟训练，我们以`dnn`为例介绍PaddleRec的简单使用。
+
+#### 单机训练
+
+```bash
+# 使用CPU进行单机训练
+python -m paddlerec.run -m paddlerec.models.rank.dnn -d cpu -e single 
+
+# 使用GPU进行单机训练
+python -m paddlerec.run -m paddlerec.models.rank.dnn -d gpu -e single
+```
+
+#### 本地模拟分布式训练
+
+```bash
+# 使用CPU资源进行本地模拟分布式训练
+python -m paddlerec.run -m paddlerec.models.rank.dnn -e local_cluster
+```
+
+#### 集群分布式训练
+
+```bash
+# 配置好 mpi/k8s/paddlecloud集群环境后
+python -m paddlerec.run -m paddlerec.models.rank.dnn -e cluster
+```
+
+### 启动内置模型的自定配置
+
+若您复用内置模型，对**yaml**配置文件进行了修改，如更改超参，重新配置数据后，可以直接使用paddlerec运行该yaml文件。
+
+我们以dnn模型为例，在paddlerec代码目录下，修改了dnn模型`config.yaml`的配置后，运行`dnn`模型：
+```bash
+python -m paddlerec.run -m ./models/rank/dnn/config.yaml -e single
+```
+
+
+<h2 align="center">支持模型列表</h2>
+
+
+|   方向   |                                      模型                                      | 单机CPU训练 | 单机GPU训练 | 分布式CPU训练 |
+| :------: | :----------------------------------------------------------------------------: | :---------: | :---------: | :-----------: |
+| 内容理解 | [Text-Classifcation](models/contentunderstanding/classification/model.py) |      ✓      |      x      |       ✓       |
+| 内容理解 |           [TagSpace](models/contentunderstanding/tagspace/model.py)            |      ✓      |      x      |       ✓       |
+|   召回   |                      [TDM](models/treebased/tdm/model.py)                      |      ✓      |      x      |       ✓       |
+|   召回   |                  [Word2Vec](models/recall/word2vec/model.py)                   |      ✓      |      x      |       ✓       |
+|   召回   |                       [SSR](models/recall/ssr/model.py)                        |      ✓      |      ✓      |       ✓       |
+|   召回   |                   [Gru4Rec](models/recall/gru4rec/model.py)                    |      ✓      |      ✓      |       ✓       |
+|   排序   |                        [Dnn](models/rank/dnn/model.py)                         |      ✓      |      x      |       ✓       |
+|   排序   |                     [DeepFM](models/rank/deepfm/model.py)                      |      ✓      |      x      |       ✓       |
+|   排序   |                    [xDeepFM](models/rank/xdeepfm/model.py)                     |      ✓      |      x      |       ✓       |
+|   排序   |                        [DIN](models/rank/din/model.py)                         |      ✓      |      x      |       ✓       |
+|   排序   |                  [Wide&Deep](models/rank/wide_deep/model.py)                   |      ✓      |      x      |       ✓       |
+|  多任务  |                     [ESMM](models/multitask/esmm/model.py)                     |      ✓      |      ✓      |       ✓       |
+|  多任务  |                     [MMOE](models/multitask/mmoe/model.py)                     |      ✓      |      ✓      |       ✓       |
+|  多任务  |             [ShareBottom](models/multitask/share-bottom/model.py)              |      ✓      |      ✓      |       ✓       |
+|   匹配   |                       [DSSM](models/match/dssm/model.py)                       |      ✓      |      x      |       ✓       |
+|   匹配   |           [MultiView-Simnet](models/match/multiview-simnet/model.py)           |      ✓      |      x      |       ✓       |
+
+
+
+<h2 align="center">文档</h2>
+
+### 背景介绍
+* [推荐系统介绍](doc/rec_background.md)
+* [分布式深度学习介绍](doc/ps_background.md)
+
+### 新手教程
+* [环境要求](#环境要求)
+* [安装命令](#安装命令)
+* [快速开始](#启动内置模型的默认配置)
+
+### 进阶教程
+* [自定义数据集及Reader](doc/custom_dataset_reader.md)
+* [分布式训练](doc/distributed_train.md)
+
+### 开发者教程
+* [PaddleRec设计文档](doc/design.md)
+
+### 关于PaddleRec性能
+* [Benchmark](doc/benchmark.md)
+
+### FAQ
+* [常见问题FAQ](doc/faq.md)
+
+
+<h2 align="center">社区</h2>
+
+### 反馈
+如有意见、建议及使用中的BUG，欢迎在`GitHub Issue`提交
+
+### 版本历史
+- 2020.5.14 - PaddleRec v0.1
+  
+### 许可证书
+本项目的发布受[Apache 2.0 license](LICENSE)许可认证。
+  
+>>>>>>> d7171ec5daa477584de89ea7e57a382045e12311