Deep Learning 101 using Paddle

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

Deep Learning with PaddlePaddle

• Fit a Line
• Recognize Digits
• Image Classification
• Word to Vector
• Recommender System
• Understand Sentiment
• Label Semantic Roles
• Machine Translation
  

Running the Book

This book you are reading is interactive -- each chapter can run as a Jupyter Notebook.

We packed this book, Jupyter, PaddlePaddle, and all dependencies into a Docker image. So you don't need to install anything except Docker. If you are using Windows, please follow this installation guide. If you are running Mac, please follow this. For various Linux distros, please refer to https://www.docker.com. If you are using Windows or Mac, you might want to give Docker more memory and CPUs/cores.

Just type

docker run -d -p 8888:8888 paddlepaddle/book

This command will download the pre-built Docker image from DockerHub.com and run it in a container. Please direct your Web browser to http://localhost:8888 to read the book.

If you are living in somewhere slow to access DockerHub.com, you might try our mirror server docker.paddlepaddle.org:

docker run -d -p 8888:8888 docker.paddlepaddle.org/book
Training with GPU

By default we are using CPU for training, if you want to train with GPU, the steps are a little different.

To make sure GPU can be successfully used from inside container, please install nvidia-docker. Then run:

nvidia-docker run -d -p 8888:8888 paddlepaddle/book:latest-gpu

Or you can use the image registry mirror in China:

nvidia-docker run -d -p 8888:8888 docker.paddlepaddle.org/book:latest-gpu

Change the code in the chapter that you are reading from

paddle.init(use_gpu=False, trainer_count=1)

to:

paddle.init(use_gpu=True, trainer_count=1)
Contribute

Your contribution is welcome! Please feel free to file Pull Requests to add your chapter as a directory under /pending. Once it is going stable, the community would like to move it to /.

To write, run, and debug your chapters, you will need Python 2.x, Go >1.5. You can build the Docker image using this script. This tutorial is contributed by PaddlePaddle, and licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

本帖隐藏的内容

Deep Learning 101 Using Paddle.pdf (18.42 MB)

2017-9-20 03:35:44 上传

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏18 回帖

1. VGG
   
2. 
   
3. Let’s start with the VGG model. Since the image size and amount of CIFAR10 are relatively small comparing to ImageNet, we use a small version of VGG network for CIFAR10. Convolution groups incorporate BN and dropout operations.
   
4. 
   
5. Define input data and its dimension
   
6. 
   
7. The input to the network is defined as paddle.layer.data, or image pixels in the context of image classification. The images in CIFAR10 are 32x32 color images of three channels. Therefore, the size of the input data is 3072 (3x32x32), and the number of categories is 10.
   
8. 
   
9. datadim = 3 * 32 * 32
   
10. classdim = 10
    
11. image = paddle.layer.data(
    
12.      name="image", type=paddle.data_type.dense_vector(datadim))

复制代码

1. Define VGG main module
   
2. 
   
3. net = vgg_bn_drop(image)
   
4. The input to VGG main module is from the data layer. vgg_bn_drop defines a 16-layer VGG network, with each convolutional layer followed by BN and dropout layers. Here is the definition in detail:
   
5. 
   
6. def vgg_bn_drop(input):
   
7.      def conv_block(ipt, num_filter, groups, dropouts, num_channels=None):
   
8.          return paddle.networks.img_conv_group(
   
9.              input=ipt,
   
10.              num_channels=num_channels,
    
11.              pool_size=2,
    
12.              pool_stride=2,
    
13.              conv_num_filter=[num_filter] * groups,
    
14.              conv_filter_size=3,
    
15.              conv_act=paddle.activation.Relu(),
    
16.              conv_with_batchnorm=True,
    
17.              conv_batchnorm_drop_rate=dropouts,
    
18.              pool_type=paddle.pooling.Max())
    
19. 
    
20.      conv1 = conv_block(input, 64, 2, [0.3, 0], 3)
    
21.      conv2 = conv_block(conv1, 128, 2, [0.4, 0])
    
22.      conv3 = conv_block(conv2, 256, 3, [0.4, 0.4, 0])
    
23.      conv4 = conv_block(conv3, 512, 3, [0.4, 0.4, 0])
    
24.      conv5 = conv_block(conv4, 512, 3, [0.4, 0.4, 0])
    
25. 
    
26.      drop = paddle.layer.dropout(input=conv5, dropout_rate=0.5)
    
27.      fc1 = paddle.layer.fc(input=drop, size=512, act=paddle.activation.Linear())
    
28.      bn = paddle.layer.batch_norm(
    
29.          input=fc1,
    
30.          act=paddle.activation.Relu(),
    
31.          layer_attr=paddle.attr.Extra(drop_rate=0.5))
    
32.      fc2 = paddle.layer.fc(input=bn, size=512, act=paddle.activation.Linear())
    
33.      return fc2

复制代码

Paddle是个不错的计量经济学软件

Thanks 4 sharing！

{:3_50:}