Diverse Application Scenarios

PaddlePaddle powers hundreds of Baidu's and its partners' products.


Computer Vision

Using a Convolutional Neural Network (CNN) for Image Recognition and Object Detection.


Natural Language Understanding

Using a Recurrent Neural Network (RNN) for Sentiment Analysis.


Recommender System

Using Deep Learning on Recommendation Systems to Help Users Find Interests.


Quick Start

Install and make predictions in 5 minutes

Download the trained housing prices model and then install PaddlePaddle on your computer:

pip install paddlepaddle

Instructions on installing pip

Now, create a new file called housing.py, and paste this Python code (make sure to set the right path based on the location of fit_a_line.tar on your computer):

import paddle.v2 as paddle

# Initialize PaddlePaddle.
paddle.init(use_gpu=False, trainer_count=1)

# Configure the neural network.
x = paddle.layer.data(name='x', type=paddle.data_type.dense_vector(13))
y_predict = paddle.layer.fc(input=x, size=1, act=paddle.activation.Linear())

with open('fit_a_line.tar', 'r') as f:
    parameters = paddle.parameters.Parameters.from_tar(f)

# Infer using provided test data.
probs = paddle.infer(
     output_layer=y_predict, parameters=parameters,
     input=[item for item in paddle.dataset.uci_housing.test()()])

for i in xrange(len(probs)):
     print 'Predicted price: ${:,.2f}'.format(probs[i][0] * 1000)
        

Run python housing.py and voila! It should print out a list of predictions for the test housing data.

Technical Features

Ease of use

PaddlePaddle provides an intuitive and flexible interface for loading data and specifying model structures.

Flexibility

It supports CNN, RNN and various variants and configures complicated deep models easily.

Efficiency

It also provides extremely optimized operations, memory recycling, and network communication.

Scalability

PaddlePaddle makes it easy to scale heterogeneous computing resources and storage to accelerate the training process.