Fast Machine Learning Lab

hls4ml

Machine learning on FPGAs using HLS

hls4ml-tutorial

Tutorial notebooks for hls4ml

qonnx

QONNX: Arbitrary-Precision Quantized Neural Networks in ONNX

example-models

qonnx_model_zoo

Model zoo for the Quantized ONNX (QONNX) model format

hls4ml-live-demo

Live demo of hls4ml on embedded platforms such as the Pynq-Z2

l1-jet-id

Training, testing and synthetisation for FPGA of three machine learning models: MLP, deepsets, and interaction networks with application to HL-LHC jet tagging.

SuperSONIC

Server infrastructure for GPU inference-as-a-service in large scientific experiments

nac-opt

Neural Architecture Codesign: A Model Optimization Toolkit for Physics

ml4fg

Machine Learning on frame grabbers for ultra-low latency in situ inference

fastmachinelearning.github.io

sonic-workflows

hawq-jet-tagging

ml-quantum-readout

db4hls

wa-hls4ml

build_triton

Build Triton inference server container for CMS

cloud-segmentation

cmssw

CMS Offline Software

iccad2023

Fast Machine Learning for Science Workshop @ ICCAD 2023

onnxruntime_backend

The Triton backend for the ONNX Runtime.

physical-cartpole

Work related to the Inverted Pendulum Hardware

pose

11ty static site

pytorch_backend

The Triton backend for the PyTorch TorchScript models.

qkeras

QKeras: a quantization deep learning library for Tensorflow Keras

RecoBTag-Combined

RecoBTag-Combined

RN0X_Pokemon

Tiny, Quantized Neural Network, orignially based on ResNet8, trained to recognize pokemon

server

The Triton Inference Server provides an optimized cloud and edge inferencing solution.

smart-pixels

wa-hls4ml-paper

Code for plots, models, data generation and other utilities relating to the paper "wa-hls4ml: A Benchmark and Surrogate Models for hls4ml Resource and Latency Estimation"