lhz1029

t5patches

T5Patches is a set of tools for fast and targeted editing of generative language models built with T5X.

cs181-demos

IPython demos for CS181

Enso

Enso: An Open Source Library for Benchmarking Embeddings + Transfer Learning Methods

finetune

Scikit-learn style model finetuning for NLP

autoencoding_vi_for_topic_models

Tensorflow implementation for prodLDA and NVLDA.

Beta-VAE

Pytorch implementation of β-VAE

fakeR

An R package to generate fake data given an existing dataset

lhz1029.github.io

deep_permutation_invariant

Code accompanying paper "Set Norm and Equivariant Skip Connections: Putting the Deep in Deep Sets."

nuisance-aware-ood-detection

Code for paper "Robustness to Spurious Correlations Improves Semantic Out-of-Distribution Detection"

lhz1029/fakeR

An R package to generate fake data given an existing dataset

lhz1029/lhz1029.github.io

lhz1029/autoencoding_vi_for_topic_models

Tensorflow implementation for prodLDA and NVLDA.

lhz1029/Beta-VAE

Pytorch implementation of β-VAE

lhz1029/breakout-rl

Playing Atari Breakout with Q-learning

lhz1029/deep_Mahalanobis_detector

Code for the paper "A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks".

lhz1029/DenseNet

DenseNet implementation in Keras

lhz1029/direct-preference-optimization

Reference implementation for DPO (Direct Preference Optimization)

lhz1029/distribution_augmentation

Code for the paper, "Distribution Augmentation for Generative Modeling", ICML 2020.

lhz1029/dlcl

The implementation of "Learning Deep Transformer Models for Machine Translation"

lhz1029/DomainBed

DomainBed is a suite to test domain generalization algorithms

lhz1029/generative_models

Pytorch implementations of generative models: VQVAE2, AIR, DRAW, InfoGAN, DCGAN, SSVAE

lhz1029/glow

Code for reproducing results in "Glow: Generative Flow with Invertible 1x1 Convolutions"

lhz1029/google-research

Google Research

lhz1029/jax

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

lhz1029/Khan_Academy_Infection

A Khan Academy Project-based Interview Question

lhz1029/learning-with-preferences

Train transformer language models with preference data.

lhz1029/manifold-flow

Manifold-learning flows (ℳ-flows)

lhz1029/MD_CA_legislation

A quick analysis on legislation proposed in previous sessions in Maryland and California.

lhz1029/pixel-cnn

Code for the paper "PixelCNN++: A PixelCNN Implementation with Discretized Logistic Mixture Likelihood and Other Modifications"

lhz1029/Premise_data_challenge

Analyzing food price data collected from on-the-ground contributors.

lhz1029/Spurious_OOD

lhz1029/t5patches

T5Patches is a set of tools for fast and targeted editing of generative language models built with T5X.

lhz1029/uncertainty-baselines

High-quality implementations of standard and SOTA methods on a variety of tasks.

lhz1029/WellWatchers

Predicting the status of wells across Tanzania

lhz1029/wilds

A machine learning benchmark of in-the-wild distribution shifts, with data loaders, evaluators, and default models.