Skip to content
This repository is a tutorial targeting how to train a deep neural network model in a higher efficient way. In this repository, we focus on two main frameworks that are Keras and Tensorflow.
Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
keras_basis
keras_multiple_gpus
tensorflow_basis
tensorflow_cluster
tensorflow_multiple_gpus
.gitignore
README.md

README.md

Model Training in Keras and Tensorflow

This repository is a tutorial targeting how to train a deep neural network model in a higher efficient way. In this repository, we focus on two main frameworks that are Keras and Tensorflow. Parts of the content (scripts, documents, etc.) were referred from Tensorflow/Model repository and Keras document. The relative framework version please refer to the top description of each script. Notice the backend of Keras is Tensorflow in this repository.

The document to this repository :

The following are the list for the main reference.

Content

At the beginning of each framework, we introduced the basic model training process and its implementation script as well. After that, we introduced two different types of advanced model training that are (1) in multiple graphics cards (GPUs) and (2) in multiple machines/containers with multiple GPUs.

We used CIFAR-10 image dataset and tried to train a deep neural network model recognizing them (a typical classification problem) as an example.

Keras

Tensorflow

You can’t perform that action at this time.