Basic Deep Learning Models

This section collects 8 documents arranging from the simplest to the most challenging, which will guide you through the basic deep learning tasks in PaddlePaddle.

The documentation in this chapter covers a lot of deep learning basics and how to implement them with PaddlePaddle. See the instructions below for how to use:


The book you are reading is an “interactive” e-book - each chapter can be run in a Jupyter Notebook.

We packaged Jupyter, PaddlePaddle, and various dependency softwares into a Docker image. It frees you from installing these softwares by yourself, and you only need to just install Docker. For various Linux versions, please refer to . If you use docker on Windows or Mac , consider allocate more Memory and CPU resources to Docker .


This book assumes you are performing CPU training by default. If you want to use GPU training, the steps will vary slightly. Please refer to “GPU Training” below.

CPU training

Just run these in shell:

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

It downloads the Docker image for running books from To read and edit this book on-line, please visit http://localhost:8888 in your browser.

If the Internet connection to is compromised, try our spare docker image named

docker run -d -p 8888:8888

GPU training

To ensure that the GPU driver works properly in the image, we recommend running the image with nvidia docker . Please install nvidia-docker first, then run:

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

Or use a image source in China to run:

nvidia-docker run -d -p 8888:8888

modify the following codes

use_cuda = False

into :

use_cuda = True

Contribute to Book

We highly appreciate your original contributions of new chapters to Book! Just Pull Requests of your contributions to the sub-directory in pending . When this chapter is endorsed, we’ll gladly move it to the root directory.

For writing, running, debugging, you need to install shell to generate Docker image。

Please Note: We also provide English Readme for PaddlePaddle book