Xilinx ML Suite |
---|
The Xilinx Machine Learning (ML) Suite provides users with the tools to develop and deploy Machine Learning applications for Real-time Inference. It provides support for many common machine learning frameworks such as Caffe, MxNet and Tensorflow as well as Python and RESTful APIs.
The ML Suite is composed of three basic parts:
- xDNN IP - High Performance general CNN processing engine.
- xfDNN Middleware - Software Library and Tools to Interface with ML Frameworks and optimize them for Real-time Inference.
- ML Framework and Open Source Support - Support for high level ML Frameworks and other open source projects.
Learn More: ML Suite Overview
Watch: Webinar on Xilinx FPGA Accelerated Inference
Getting Started
- Clone ML Suite
git clone https://github.com/Xilinx/ml-suite.git
- Install Anaconda2.
# Ensure that you ran the fix_caffe_opencv_symlink.sh script
- Install git lfs
- Go into the ml-suite directory and pull down the models
cd ml-suite; git pull
TEMPORARY NOTE:
If you are evaluating on AWS, the binaries we have included support the latest Amazon Shell
DSA name: xilinx_aws-vu9p-f1-04261818_dynamic_5_0
The Xilinx ml-suite AMI was bundled for an older shell
For this reason, if you are starting your evaluation today, it is best to begin from the FPGA Developer AMI:
If you are using the AWS EC2 F1 FPGA DEVELOPER AMI the following steps are necessary to setup the drivers:
git clone https://github.com/aws/aws-fpga.git
cd aws-fpga
source sdaccel_setup.sh
Remember that AWS requires users to run as root to control the FPGA, so the following is necessary to use Anaconda as root:
- Become root
sudo su
- Set Environment Variables Required by runtime
source <MLSUITE_ROOT>/overlaybins/setup.sh aws
- Set User Environment Variables Required to run Anaconda
source ~centos/.bashrc
- Activate the users Anaconda Virtual Environment
source activate ml-suite
You can avoid disk space problems on the FPGA DEVELOPER AMI by creating an instance with more than the default 70G of storage, or by resizing the /swapfile to something less than 35G.
Once your environment is set up, take a look at some of the command line tutorials and Jupyter Notebooks here:
Minimum System Requirements
- OS: Ubuntu 16.04.2 LTS, CentOS
- CPU: 4 Cores (Intel/AMD)
- Memory: 8 GB
Supported Platforms
Cloud Services
On Premise Platforms
- Xilinx Virtex UltraScale+ FPGA VCU1525 Acceleration Development Kit
- Note: The
xilinx_vcu1525_dynamic_5_1
DSA is required to be installed. Installation information can be found on page 118 of UG1023
- Note: The