GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → Xilinx → Chaidnn

Xilinx / Chaidnn

Licence: other
HLS based Deep Neural Network Accelerator Library for Xilinx Ultrascale+ MPSoCs

Labels

deep-neural-networksinferencednn

Projects that are alternatives of or similar to Chaidnn

Rmdl
RMDL: Random Multimodel Deep Learning for Classification
Stars: ✭ 375 (+45.35%)
Mutual labels:  deep-neural-networks, dnn
Opentpod
Open Toolkit for Painless Object Detection
Stars: ✭ 106 (-58.91%)
Mutual labels:  deep-neural-networks, dnn
Awesome Emdl
Embedded and mobile deep learning research resources
Stars: ✭ 554 (+114.73%)
Mutual labels:  deep-neural-networks, inference
Tf2
An Open Source Deep Learning Inference Engine Based on FPGA
Stars: ✭ 113 (-56.2%)
Mutual labels:  inference, dnn
Pytorch Kaldi
pytorch-kaldi is a project for developing state-of-the-art DNN/RNN hybrid speech recognition systems. The DNN part is managed by pytorch, while feature extraction, label computation, and decoding are performed with the kaldi toolkit.
Stars: ✭ 2,097 (+712.79%)
Mutual labels:  deep-neural-networks, dnn
Dancinggaga
AI 尬舞机
Stars: ✭ 315 (+22.09%)
Mutual labels:  deep-neural-networks, inference
Awesome System For Machine Learning
A curated list of research in machine learning system. I also summarize some papers if I think they are really interesting.
Stars: ✭ 1,185 (+359.3%)
Mutual labels:  deep-neural-networks, inference
Bmw Tensorflow Inference Api Gpu
This is a repository for an object detection inference API using the Tensorflow framework.
Stars: ✭ 277 (+7.36%)
Mutual labels:  deep-neural-networks, inference
Hey Jetson
Deep Learning based Automatic Speech Recognition with attention for the Nvidia Jetson.
Stars: ✭ 161 (-37.6%)
Mutual labels:  deep-neural-networks, inference
Monet
MONeT framework for reducing memory consumption of DNN training
Stars: ✭ 126 (-51.16%)
Mutual labels:  deep-neural-networks, dnn
Opencv Mtcnn
An implementation of MTCNN Face detector using OpenCV's DNN module
Stars: ✭ 59 (-77.13%)
Mutual labels:  inference, dnn
Adversarial Robustness Toolbox
Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams
Stars: ✭ 2,638 (+922.48%)
Mutual labels:  deep-neural-networks, inference
Mobilnet ssd opencv
MobilNet-SSD object detection in opencv 3.4.1
Stars: ✭ 64 (-75.19%)
Mutual labels:  deep-neural-networks, dnn
Onnx
Open standard for machine learning interoperability
Stars: ✭ 11,829 (+4484.88%)
Mutual labels:  deep-neural-networks, dnn
Bmw Yolov4 Inference Api Cpu
This is a repository for an nocode object detection inference API using the Yolov4 and Yolov3 Opencv.
Stars: ✭ 180 (-30.23%)
Mutual labels:  deep-neural-networks, inference
Models
Model Zoo for Intel® Architecture: contains Intel optimizations for running deep learning workloads on Intel® Xeon® Scalable processors
Stars: ✭ 248 (-3.88%)
Mutual labels:  deep-neural-networks, inference
JOCI
Ordinal Common-sense Inference
Stars: ✭ 27 (-89.53%)
Mutual labels:  inference
rindow-neuralnetworks
Neural networks library for machine learning on PHP
Stars: ✭ 37 (-85.66%)
Mutual labels:  dnn
dnn
A light-weight deep learning framework implemented in C++.
Stars: ✭ 12 (-95.35%)
Mutual labels:  dnn
speech-to-text
mixlingual speech recognition system; hybrid (GMM+NNet) model; Kaldi + Keras
Stars: ✭ 61 (-76.36%)
Mutual labels:  dnn
View All Similar Projects ➔

CHaiDNN-v2

Analysis and Eval
Supported Layers Performance/Resource Utilization
Performance Eval
Design and Development
API Reference Quantization User Guide for CHaiDNN Model Zoo Running Inference on new Network
Creating SDx GUI Project Configurable Parameters Custom Platform Generation Software Layer Plugin
SDSoC Environment User Guide Hardware-Software Partitioning for Performance

Introduction

CHaiDNN is a Xilinx Deep Neural Network library for acceleration of deep neural networks on Xilinx UltraScale MPSoCs. It is designed for maximum compute efficiency at 6-bit integer data type. It also supports 8-bit integer data type.

The design goal of CHaiDNN is to achieve best accuracy with maximum performance. The inference on CHaiDNN works in fixed point domain for better performance. All the feature maps and trained parameters are converted from single precision to fixed point based on the precision parameters specified by the user. The precision parameters can vary a lot depending upon the network, datasets, or even across layers in the same network. Accuracy of a network depends on the precision parameters used to represent the feature maps and trained parameters. Well-crafted precision parameters are expected to give accuracy similar to accuracy obtained from a single precision model.

What's new in CHaiDNN-v2

  • 4x GOPS compared to CHaiDNN-v1 (2017.4) (Performance numbers)

  • 2x MAC on DSPs at int6

  • Double-Pumped DSPs allowing the DSPs to be clocked at twice the core clock (Some configs can go upto 350/700Mhz)

  • Introducing DietChai - A miniature version of CHai for smaller MPSoC/ Zynq devices

  • 128, 256, 512, 1024 DSP design configs verified for ZU9

  • Support for URAM

  • 128, 256, 512 DSP configs verified for ZU7

  • ModelZoo of 6 networks at int8 and int6 precision

  • Support for two quantization modes - Dynamic fixed point and Xilinx Quantizer

  • Enhanced API to enable better hardware- software partitioning for users

  • Support for software custom layer plug-ins

  • Fully Connected layers on CPU

  • More documentation

Performance Benchmarks(fps)

Network Xilinx CHai w/ 1024DSP @ 250/500MHz (Measured on ZU9) Nvidia Jetson TX2 @ 1.3GHz*
GoogleNet-6bit w/o FC 220 Googlenet-16FP: 201
GoogleNet-6bit w/ FC 207
GoogleNet-8bit w/o FC 151
GoogleNet-8bit w/ FC 145
Alexnet-6bit w/o FC 606 Alexnet-16FP: 250
Alexnet-6bit w/ FC 10
Alexnet-8bit w/o FC 390
Alexnet-8bit w/ FC 10

* Source: https://devblogs.nvidia.com/jetson-tx2-delivers-twice-intelligence-edge/

Hardware and Software Requirements

The CHaiDNN library is designed to work with Zynq UltraScale+ MPSoCs. The library has been verified on zcu102 and zcu104 boards. Xilinx SDSoC 2018.2 Development Environment is required to work with the library.

How to Download the Repository

To get a local copy of the CHaiDNN repository, configure git-lfs and then, clone this repository to the local system with the following command:

git clone https://github.com/Xilinx/CHaiDNN.git CHaiDNN

Where CHaiDNN is the name of the directory where the repository will be stored on the local system. This command needs to be executed only once to retrieve the latest version of the CHaiDNN library.

GitHub Repository Structure 
CHaiDNN/
|
|-- CONTRIBUTING.md
|-- LICENSE
|-- README.md
|-- SD_Card
|   |-- lib
|   |-- cblas
|   |-- images
|   |-- opencv
|   |-- protobuf
|   |-- zcu102
|   `-- zcu104
|-- design
|   |-- build
|   |-- conv
|   |-- deconv
|   |-- pool
|   `-- wrapper
|-- docs
|   |-- API.md
|   |-- BUILD_USING_SDX_GUI.md
|   |-- CONFIGURABLE_PARAMS.md
|   |-- CUSTOM_PLATFORM_GEN.md
|   |-- HW_SW_PARTITIONING.md
|   |-- MODELZOO.md
|   |-- PERFORMANCE_SNAPSHOT.md
|   |-- QUANTIZATION.md
|   |-- RUN_NEW_NETWORK.md
|   |-- SOFTWARE_LAYER_PLUGIN.md
|   |-- SUPPORTED_LAYERS.md
|   `-- images
|-- software
|   |-- bufmgmt
|   |-- checkers
|   |-- common
|   |-- custom
|   |-- example
|   |-- imageread
|   |-- include
|   |-- init
|   |-- interface
|   |-- scheduler
|   |-- scripts
|   |-- swkernels
|   `-- xtract
`-- tools
    |-- SETUP_TOOLS.md
    `-- tools.zip

Run Inference

Using Pre-built binaries

To run inference on example networks, follow these steps:

  1. Download the example network 6-bit GoogleNet with Xilinx Quantization scheme. More networks are available as part of the ModelZoo.

  2. Place the downloaded and unzipped contents at "SD_Card/models" directory. Create SD_Card/models directory if not present already.

  3. Copy the required contents of "SD_Card" folder into a SD-Card.

    • opencv
    • protobuf
    • cblas
    • images
    • bit-stream, boot loader, lib & executables (either from SD_Card/zcu102 or SD_Card/zcu104)

  4. Insert the SD-Card and power ON the board.

    📌 NOTE: A serial port emulator (Teraterm/Minicom) is required to interface the user commands to the board

  5. Attach a USB-UART cable from the board to the host PC. Set the UART serial port to

    Baud rate: 115200
Data: 8 bit
Parity: none
Stop: 1 bit
Flow control: none

  6. After boot sequence, set LD_LIBRARY_PATH env variable.

    export OPENBLAS_NUM_THREADS=2
export LD_LIBRARY_PATH=lib/:opencv/arm64/lib/:protobuf/arm64/lib:cblas/arm64/lib

  7. Create a folder "out" inside the network directory to save the outputs sh cd /mnt mkdir models/<network>/out

  8. Execute "*.elf" file to run inference

    • The format for running these example networks is described below: 
      ./<example network>.elf <quantization scheme> <bit width> <img1_path> <img2_path>
    • For GoogleNet 6-bit inference with Xilinx quantization scheme execute the following 
      ./googlenet.elf Xilinx 6 images/camel.jpg images/goldfish.JPEG

  9. Sync after execution

    cd /
sync
umount /mnt

  10. Output will be written into text file inside respective output folders.

    Ex : models/<network>/out

📌 NOTE: Failing to run sync might corrupt the file system and cause crash on subsequent runs.

📌 NOTE: For running inference on a new network, please follow the instructions in Run new Network using CHaiDNN.

Build from Source

CHaiDNN can be built using Makefiles OR using SDx IDE. The below steps describe how to build CHaiDNN using Makefiles. For steps to build using SDx IDE, see the instructions in Build using SDx IDE.

Build CHaiDNN Hardware

Please follow the steps to build the design for zcu102 (ZU9 device based board)

  1. Please generate a custom platform with 1x and 2x clocks using the steps described here. With Chai-v2, we now have the DSPs operating at twice the frequency of the rest of the core.

  2. Go to CHaiDNN/design/build folder.

  3. Set SDx tool environment

    • For BASH: 
      source <SDx Installation Dir>/installs/lin64/SDx/2018.2/settings64.sh
    • For CSH 
      source <SDx Installation Dir>/installs/lin64/SDx/2018.2/settings64.csh

  4. To build the design, run Makefile. (By default this will build 1024 DSP design @ 200/400 MHz)

    make ultraclean
make

    📌 NOTE:

    • To build DietChai, run make DIET_CHAI_Z=1. This builds a design with 128 compute DSPs and 64-bit AXI interface. Run make DIET_CHAI_ZUPLUS=1 to build a design with 128 compute DSPs and 128-bit AXI interface.
    • To exclude deconv Kernel, set DECONV_ENABLE=0 in Makefile. Default is DECONV_ENABLE=1.
    • To exclude Pool Kernel, set POOL_ENABLE=0 in Makefile. With this setting, Pooling functionality embedded in Convolution accelerator is used. Default is POOL_ENABLE=1.
    • When building DietChai, do not change POOL_ENABLE, DECONV_ENABLE values in Makefile.

  5. After the build is completed, copy the libxlnxdnn.so file and other build files (BOOT.BIN, image.ub and _sds directory) inside build/sd_card to SD_Card directory.

    make copy

  6. The hardware setup is now ready.

📌 NOTE:

  • The 1024 DSP config was timing closed at 250/500Mhz with an iterative synthesis and P&R strategy. In the first iteration, the design was taken through the SDx flow (all the way till the bitstream Gen) at 200/400Mhz. In the second iteration the post-routed design from the first iteration was re-routed at 250/500Mhz. We believe that this is a general strategy that can be applied for other configs also. We would definitely like to hear from you on this if you are able to crank the frequency further up on other configs with this strategy.
  • Please note that when you try building some of the configs that are mentioned in the performance table, you might see some negative slack reported by the tools but we encourage you to try the bitstreams generated on hardware for functionality. These timing closure issues can be cleaned up with some special synthesis and P&R strategies. (You are welcome to try the timing-closure strategies that have worked for you in the past on other designs.)
Build CHaiDNN Software

Follow the steps to compile the software stack.

  1. Copy libxlnxdnn.so to SD_Card/lib directory. The libxlnxdnn.so file can be found in the design/build/sd_card directory once the HW build is finished. You can skip this step if have already copied the libxlnxdnn.so file to the suggested directory.

  2. Set the SDx tool environment.

    • CSH 
      source <SDx Installation Dir>/installs/lin64/SDx/2018.2/settings64.csh
    • BASH 
      source <SDx Installation Dir>/installs/lin64/SDx/2018.2/settings64.sh

  3. Go to the software directory. This contains all the files to generate software libraries (.so).

    cd <path to CHaiDNN>/software

  4. Go to scripts directory, open Makefile and update the SDx_BUILD_PATH variable. See example below.

    SDx_BUILD_PATH = <SDx Installation Dir>/installs/lin64/SDx/2018.2

  5. Now run the following commands.

    make ultraclean
make

    📌 NOTE:

    • To build DietChai, run make DIET_CHAI_Z=1. This builds a design with 128 compute DSPs and 64-bit AXI interface. Run make DIET_CHAI_ZUPLUS=1 to build a design with 128 compute DSPs and 128-bit AXI interface.
    • To exclude deconv Kernel, set DECONV_ENABLE=0 in Makefile. Default is DECONV_ENABLE=1.
    • To exclude Pool Kernel, set POOL_ENABLE=0 in Makefile. With this setting, Pooling functionality embedded in Convolution accelerator is used. Default is POOL_ENABLE=1.
    • When building DietChai, do not change POOL_ENABLE, DECONV_ENABLE values in Makefile.

    📌 NOTE: Ensure that the software and the hardware are build with the same settings.

  6. Make will copy all executables to SD_Card directory and all .so files to SD_Card/lib directory.

  7. Now, we are set to run inference. Follow the steps mentioned in "run inference using pre-build binaries"

Additional Resources and Support

For questions and to get help on this project or your own projects, visit the CHaiDNN Github Issues.

License and Contributing to the Repository

The source for this project is licensed under the Apache License 2.0

To contribute to this project, follow the guidelines in the Repository Contribution README

Acknowledgements
Revision History
Date Readme Version Release Notes Tool Version
Feb, 2018 1.0 Initial Xilinx release SDx-2017.4
June, 2018 2.0 CHaiDNN-v2 SDx-2018.2
Deprecated Features
  • 16-bit activations

Copyright© 2018 Xilinx

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].