jaeyong-song
Deep Learning and System
Grad Student@AISys, Seoul National University Gwangmyeong-si, Korea
Pinned Repositories
GraNNDis_Artifact
[PACT'24] GraNNDis. A fast and unified distributed graph neural network (GNN) training framework for both full-batch (full-graph) and mini-batch training. Provides unification of full-/mini-batch training using a novel data/communication structure.
Optimus-CC
[ASPLOS'23] Optimus-CC: Efficient Large NLP Model Training with 3D Parallelism Aware Communication Compression
algarve_attractions_crawl_clustering
Get information of algarve(faro) attractions from google map and make n-cluster (for getting main positions).
chinground
친구를 만나는 운동장 친그라운드 / 제주 SW 개발자 양성과정 / LIKELION in Jeju / 멋쟁이 사자처럼 in 제주
CSR_converter
cuda_sorting_algo
Cuda exact and quasi sorting implementations
deep-gradient-compression
[ICLR 2018] Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
DSL_EDA_jeju_jaenan
제주재난지원금 데이터 EDA 프로젝트. 연세대학교 데이터사이언스랩 2021_1 EDA 프로젝트
jaeyong-song
Config files for my GitHub profile.
jaeyong-song's Repositories
jaeyong-song/DSL_EDA_jeju_jaenan
제주재난지원금 데이터 EDA 프로젝트. 연세대학교 데이터사이언스랩 2021_1 EDA 프로젝트
jaeyong-song/jaeyong-song
Config files for my GitHub profile.
jaeyong-song/algarve_attractions_crawl_clustering
Get information of algarve(faro) attractions from google map and make n-cluster (for getting main positions).
jaeyong-song/CSR_converter
jaeyong-song/cuda_sorting_algo
Cuda exact and quasi sorting implementations
jaeyong-song/deep-gradient-compression
[ICLR 2018] Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
jaeyong-song/deep_learning_advanced
Deep Learning study with advanced (SOTA) models and skills.
jaeyong-song/dgl_large_synthetic_graph
Large synthetic graph dataset generator for DGL.
jaeyong-song/distbelief
Implementing Google's DistBelief paper
jaeyong-song/DSL_lol_project
롤 경기 데이터 기반 실시간 승률 예측 프로젝트
jaeyong-song/dsl_cnn_examples
데이터사이언스랩 CNN 실습
jaeyong-song/GraNNDis_Artifact
Artifact of "GraNNDis: Fast Distributed Graph Neural Network Training Framework for Multi-Server Clusters" (PACT'24). Provides fast distributed full-batch GNN training throughput and GraNNDis optimizations.
jaeyong-song/HLS_RTL_PYNQ
Logic Design Study with HLS, RTL and PYNQ FPGA board.
jaeyong-song/hotel_booking_demand_EDA
DSL hotel booking demand EDA project
jaeyong-song/jaeyong-song.github.io
jaeyong-song/jekyll-theme-chirpy
A minimal, sidebar, responsive web design Jekyll theme that focuses on text presentation.
jaeyong-song/lg_ai_display_pj
jaeyong-song/lms_db_pj
2019-2 데이터베이스 프로젝트: LMS DB 프로젝트
jaeyong-song/machine_learning_graph
Machine Learning with Graphs(Networks) (CS224W)
jaeyong-song/machine_learning_spark
Machine Learning (w/o deep learning) study with PySpark based on CS246 and DSL examples.
jaeyong-song/network_distributed_pytorch
Internel / 1Gb / 10Gb / 100Gb distributed learning experiment with PyTorch.
jaeyong-song/oop_pj_chess
2019-2 OOP Term Project: Java Swing Chess
jaeyong-song/Optimus-CC
[ASPLOS'23] Optimus-CC: Efficient Large NLP Model Training with 3D Parallelism Aware Communication Compression
jaeyong-song/OptML_course
EPFL Course - Optimization for Machine Learning - CS-439
jaeyong-song/pinsage
pinsage for wine recommendation
jaeyong-song/powersgd
Practical low-rank gradient compression for distributed optimization: https://arxiv.org/abs/1905.13727
jaeyong-song/PyTorch
Deep Learning Zero to All - Pytorch
jaeyong-song/pytorch-cifar
95.16% on CIFAR10 with PyTorch
jaeyong-song/pytorch-classification
Classification with PyTorch.
jaeyong-song/SC_CNN
Superchannel CNN