# doonny/PipeCNN

An OpenCL-based FPGA Accelerator for Convolutional Neural Networks

Repository: https://github.com/doonny/PipeCNN
Canonical: https://ross.abutalabs.com/products/pipecnn
Language: C
License: Apache-2.0
License Family: permissive
Topics: opencl, fpga-accelerator, hls, hardware, altera-opencl-sdk, fpga, deep-learning, deep-neural-networks
Last push: 2022-02-14T23:59:36+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3586, "days_push": 1661, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1386, forks 373 (observed 2026-08-28T04:04:35.014377+00:00)

## What it is
PipeCNN is an OpenCL-based FPGA accelerator for large-scale convolutional neural network inference, written in C with pipelined kernels. It supports both Intel OpenCL SDK and Xilinx Vitis flows and includes a ModelZoo with pre-quantized VGG-16 and ResNet-50 models.

## Use cases
- accelerate cnn inference on fpga hardware
- run vgg-16 or resnet-50 image classification on an fpga board
- benchmark quantized deep neural networks on intel arria-10 or xilinx u50 cards
- learn opencl-based high-level synthesis for deep learning accelerators
- measure top-1/top-5 accuracy of cnns using fpga acceleration

## When to choose
- you have a supported Intel or Xilinx FPGA board and want OpenCL-based CNN inference
- you need a scalable, resource-configurable hardware accelerator for research or education
- you want to experiment with quantized CNN models on FPGAs

## When to avoid
- you need GPU or CPU inference without FPGA hardware
- you require training support rather than inference
- you need support for modern CNN architectures beyond VGG-16 and ResNet-50
- you want an actively developed project with frequent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, llm-inference
- domain: deep-learning, hardware, computer-vision, artificial-intelligence
- platform: cross-platform
- tags: opencl, fpga-accelerator, hls, cnn-inference, altera, xilinx-vitis, hardware-acceleration, quantization, fpga

## Member repositories
- doonny/PipeCNN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:35.014377+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:39:53.829471+00:00, confidence not recorded.
  - readme: https://github.com/doonny/PipeCNN (fetched 2026-08-28T04:04:35.014377+00:00, sha 04bf547d1b58)
- Data as of 2026-08-30T08:39:29.467469+00:00.