DISCONTINUATION OF PROJECT. This project will no longer be maintained by Intel. Intel will not provide or guarantee development of or support for this project, including but not limited to, maintenance, bug fixes, new releases or updates. Patches to this project are no longer accepted by Intel. If you have an ongoing need to use this project, are interested in independently developing it, or would like to maintain patches for the community, please create your own fork of the project.
aeon
aeon is Intel Nervana's framework-independent deep learning dataloader committed to best performance. Designed for ease-of-use and extensibility.
- Supports common media types: image, video, audio. It is ready for use with classification, segmentation, localization, transcription and more.
- Loads and augments data in parallel to framework computation.
- Examples folder contains simple scripts showing how to use it.
For fast iteration and model exploration, it is already used by fastest performance deep learning framework neon.
See the new features in our latest release.
Getting Started
Installation
First grab Aeon's dependencies:
Ubuntu (release 16.04 LTS and later):
apt-get install git clang libcurl4-openssl-dev libopencv-dev libsox-dev libboost-filesystem-dev libboost-system-dev libssl-dev
For Python 3.n
apt-get install python3-dev python3-pip python3-numpy
CentOS (release 7.2 and later):
yum install epel-release
yum install git clang gcc-c++ make cmake openssl-devel opencv-devel libcurl-devel sox-devel boost-devel boost-filesystem boost-system
For Python 2.7
yum install python-pip python-devel
For Python 3.n
yum install python-pip python34-pip python34-devel python34-opencv python34-numpy
OSX:
brew tap homebrew/science
brew install opencv
brew install sox
brew install boost
Distributed Aeon
Please take a look at distributed documentation.
Code coverage
Code coverage in aeon depends on llvm-cov and lcov.
Report will be generated in html-coverage-report/index.html
Example:
sudo apt-get install llvm lcov
mkdir build
cd build
# COVERAGE flag must be set to generate coverage report
cmake .. -DCOVERAGE=ON
# If you want to generate report when unit test fails: make -i coverage
make coverage
To install Aeon:
git clone https://github.com/NervanaSystems/aeon.git
For Python 2.7
cd aeon
pip install -r requirements.txt
mkdir -p build && cd $_ && cmake .. && pip install .
For Python 3.n
cd aeon
pip3 install -r requirements.txt
mkdir -p build && cd $_ && cmake .. && pip3 install .
Note: if installing system wide (as opposed to within a virtual environment) you may need to run sudo
Now continue on to the user guide to get started using aeon. Or to the developer guide to developing custom loaders/transformers.
Documentation
The complete documentation for aeon is available here.
Support
For any bugs or feature requests please:
Search the open and closed issues list to see if we're already working on what you have uncovered. Check that your issue/request isn't framework related. File a new issue or submit a new pull request if you have some code you'd like to contribute
License
We are releasing aeon under an open source Apache 2.0 License. We welcome you to contact us with your use cases.