ribesstefano/Mapping-Multiple-LSTM-Models-on-FPGAs
Includes the SVD-based approximation algorithms for compressing deep learning models and the FPGA accelerators exploiting such approximation mechanism, as described in the paper Mapping multiple LSTM models on FPGAs.
This project helps hardware engineers and researchers working with Field-Programmable Gate Arrays (FPGAs) to efficiently run multiple deep learning models, specifically LSTMs. It takes pre-trained LSTM models and compresses them using an SVD-based approximation algorithm, then generates code for specialized FPGA hardware accelerators. The output is a highly optimized FPGA implementation capable of processing LSTM inferences much faster and with greater power efficiency.
No commits in the last 6 months.
Use this if you need to deploy several LSTM deep learning models onto FPGA hardware and require significant improvements in processing speed and resource utilization.
Not ideal if you are looking for a general-purpose deep learning deployment solution for CPUs or GPUs, or if you need to train new LSTM models.
Stars
16
Forks
4
Language
Jupyter Notebook
License
GPL-3.0
Category
Last pushed
Mar 03, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ribesstefano/Mapping-Multiple-LSTM-Models-on-FPGAs"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
fastmachinelearning/hls4ml
Machine learning on FPGAs using HLS
alibaba/TinyNeuralNetwork
TinyNeuralNetwork is an efficient and easy-to-use deep learning model compression framework.
KULeuven-MICAS/zigzag
HW Architecture-Mapping Design Space Exploration Framework for Deep Learning Accelerators
fastmachinelearning/hls4ml-tutorial
Tutorial notebooks for hls4ml
doonny/PipeCNN
An OpenCL-based FPGA Accelerator for Convolutional Neural Networks