diff --git a/README.md b/README.md
index a5f6636c3a05912e91cdb21b207288ca7b57a6e6..f7cdf634305056a5689f7739fffe913bf5f5aba0 100644
--- a/README.md
+++ b/README.md
@@ -50,7 +50,8 @@ PaddleHub以预训练模型应用为核心具备以下特点：
 
 ### 安装命令
 
-PaddlePaddle框架的安装请查阅[飞桨快速安装](https://www.paddlepaddle.org.cn/install/quick)
+在安装paddlehub之前，请先安装PaddlePaddle深度学习框架。
+PaddlePaddle深度学习框架的安装请查阅[飞桨快速安装](https://www.paddlepaddle.org.cn/install/quick)
 
 ```shell
 pip install paddlehub
diff --git a/paddlehub/autodl/DELTA/README.md b/autodl/DELTA/README.md
similarity index 84%
rename from paddlehub/autodl/DELTA/README.md
rename to autodl/DELTA/README.md
index 7b235d46f085ba05aa41ab0dabd9317b3efd09da..8620bb0dec9c0fad617abe6537f5ac93c7c503b8 100644
--- a/paddlehub/autodl/DELTA/README.md
+++ b/autodl/DELTA/README.md
@@ -1,10 +1,12 @@
+# DELTA: DEep Learning Transfer using Feature Map with Attention for Convolutional Networks
 
-# Introduction
-This page implements the [DELTA](https://arxiv.org/abs/1901.09229) algorithm in [PaddlePaddle](https://www.paddlepaddle.org.cn/install/quick).
+## Introduction
+
+This page implements the [DELTA](https://arxiv.org/abs/1901.09229) algorithm in [PaddlePaddle](https://www.paddlepaddle.org.cn).
 
 > Li, Xingjian, et al. "DELTA: Deep learning transfer using feature map with attention for convolutional networks." ICLR 2019.
 
-# Preparation of Data and Pre-trained Model
+## Preparation of Data and Pre-trained Model
 
 - Download transfer learning target datasets, like [Caltech-256](http://www.vision.caltech.edu/Image_Datasets/Caltech256/), [CUB_200_2011](http://www.vision.caltech.edu/visipedia/CUB-200-2011.html) or others. Arrange the dataset in this way:
 ```
@@ -23,7 +25,7 @@ This page implements the [DELTA](https://arxiv.org/abs/1901.09229) algorithm in
 
 - Download [the pretrained models](https://github.com/PaddlePaddle/models/tree/release/1.7/PaddleCV/image_classification#resnet-series). We give the results of ResNet-101 below.
 
-# Running Scripts
+## Running Scripts
 
 Modify `global_data_path` in `datasets/data_path` to the path root where the dataset is.
 
diff --git a/paddlehub/autodl/DELTA/args.py b/autodl/DELTA/args.py
similarity index 100%
rename from paddlehub/autodl/DELTA/args.py
rename to autodl/DELTA/args.py
diff --git a/paddlehub/autodl/DELTA/datasets/data_path.py b/autodl/DELTA/datasets/data_path.py
similarity index 100%
rename from paddlehub/autodl/DELTA/datasets/data_path.py
rename to autodl/DELTA/datasets/data_path.py
diff --git a/paddlehub/autodl/DELTA/datasets/readers.py b/autodl/DELTA/datasets/readers.py
similarity index 100%
rename from paddlehub/autodl/DELTA/datasets/readers.py
rename to autodl/DELTA/datasets/readers.py
diff --git a/paddlehub/autodl/DELTA/main.py b/autodl/DELTA/main.py
similarity index 100%
rename from paddlehub/autodl/DELTA/main.py
rename to autodl/DELTA/main.py
diff --git a/paddlehub/autodl/DELTA/models/__init__.py b/autodl/DELTA/models/__init__.py
similarity index 100%
rename from paddlehub/autodl/DELTA/models/__init__.py
rename to autodl/DELTA/models/__init__.py
diff --git a/paddlehub/autodl/DELTA/models/resnet.py b/autodl/DELTA/models/resnet.py
similarity index 100%
rename from paddlehub/autodl/DELTA/models/resnet.py
rename to autodl/DELTA/models/resnet.py
diff --git a/paddlehub/autodl/DELTA/models/resnet_vc.py b/autodl/DELTA/models/resnet_vc.py
similarity index 100%
rename from paddlehub/autodl/DELTA/models/resnet_vc.py
rename to autodl/DELTA/models/resnet_vc.py
diff --git a/paddlehub/finetune/task/base_task.py b/paddlehub/finetune/task/base_task.py
index 10d2de74c262d9d5221c389f0afffe49de74c9ad..0a118c82f1d55a61823273aea31c9556cb016e2e 100644
--- a/paddlehub/finetune/task/base_task.py
+++ b/paddlehub/finetune/task/base_task.py
@@ -832,7 +832,9 @@ class BaseTask(object):
             self.config.checkpoint_dir,
             self.exe,
             main_program=self.main_program)
-
+        # Revise max_train_steps when incremental training
+        self.max_train_steps = self.env.current_step + self.max_train_steps / self.config.num_epoch * (
+            self.config.num_epoch - self.env.current_epoch + 1)
         return is_load_successful
 
     def load_parameters(self, dirname):