finn
FPGA accelerator
Fast and scalable neural network inference framework for FPGAs.
Dataflow compiler for QNN inference on FPGAs
770 stars
32 watching
246 forks
Language: Python
last commit: 4 days ago
Linked from 2 awesome lists
compilerdataflowfpganeural-networkquantization
ly0n/awesome-robotic-tooling
aolofsson/awesome-opensource-hardwareRelated projects:
| Repository | Description | Stars |
|---|---|---|
| xilinx/logicnets | Designs and deploys neural networks integrated with Xilinx FPGAs for high-throughput applications | 83 |
 doonny/pipecnn | A tool for accelerating convolutional neural networks on Field-Programmable Gate Arrays (FPGAs) using OpenCL-based hardware design | 1,264 |
 xboot/libonnx | An onnx inference engine for embedded devices with hardware acceleration support | 589 |
 nngen/nngen | Generates hardware-specific accelerator designs for neural networks | 340 |
 nilseuropa/ros_ncnn | A ROS wrapper for NCNN, allowing developers to integrate high-performance neural network inference into ROS projects | 63 |
 ahmedfgad/numpyann | An implementation of artificial neural networks using NumPy | 98 |
 mit-han-lab/proxylessnas | Direct neural architecture search on target task and hardware for efficient model deployment | 1,429 |
| ahmedfgad/numpycnn | A Python implementation of a Convolutional Neural Network from scratch using NumPy for building CNNs from scratch | 577 |
 modern-fortran/neural-fortran | A parallel framework for building neural networks in Fortran | 411 |
| xilinx/raft | A Python toolbox providing direct access to FPGA hardware peripherals | 23 |
 nvdla/hw | The NVDLA project provides hardware designs and tools for building deep learning inference accelerators. | 1,763 |
 microsoft/deepspeed-mii | A Python library designed to accelerate model inference with high-throughput and low latency capabilities | 1,924 |
 yixuan/minidnn | A C++ library implementing deep neural networks with good performance and modularity | 399 |
 intel/neural-compressor | Tools and techniques for optimizing large language models on various frameworks and hardware platforms. | 2,257 |
 berkeleyautomation/gqcnn | Develops and analyzes convolutional neural networks for grasp quality in robotics | 316 |