Oh-hyeon Choung's repositories
MultiscaleSegmentation
Mimicking human parallel visual information processing system (mismatch penalty) for Image semantic segmentation, using multiple feature maps.
pytorch-graphNet
Pytorch implementation of Deepmind's Graph network. Original repo: https://github.com/deepmind/graph_nets.git. Original paper: https://arxiv.org/abs/1806.01261
syn-search
Optimal Drug Synthesis Condition Searching Conversational AI, using LLMs
TPD_priority_map
Psychophysics Experiment for discriminating retinotopic or non-retinotopic mapping of priority map using TPD
InvisibleInhibitions
Computational simulation for invisible retinotopic inhibition on non-retinotopic perception
SQM_discreteness
Compare discrete vs continuous networks to investigate the function of discrete temporal integration
DN_uncrowding
Can divisive normalization explain uncrowding? Let's add recurrent procedure in DN process to see if it can tackle uncrowding problem
ChEMBL_Structure_Pipeline
ChEMBL database structure pipelines
conv_arithmetic
A technical report on convolution arithmetic in the context of deep learning
cookiecutter-pytorch-pl-tb
A Cookiecutter template for pytorch-pylight-tensorboard đź’»
CVND_Exercises
Exercise notebooks for CVND.
deep-learning
Repo for the Deep Learning Nanodegree Foundations program.
dissecting_uncrowding
Codes and model results
Ohyeon5.github.io
Github pages
openbis-bode
Bode group's openbis pybis setup
Practice
Personnal practice repo for online and offline courses. HWs, mini projects, and side projects are archived here.
scikit-mol
scikit-learn classes for molecular vectorization using RDKit
scratch
Oh-hyeon's general documentation repository
segmentation
Tensorflow implementation : U-net and FCN with global convolution
tf-image-segmentation
Image Segmentation framework based on Tensorflow and TF-Slim library
tf_unet
Generic U-Net Tensorflow implementation for image segmentation
workshops
A few exercises for use at events.
zeroshot-imitation
[ICLR 2018] TensorFlow code for zero-shot visual imitation by self-supervised exploration