alexlanglang Goto Github PK

590

ahb-r-package

AHB (Adaptive Hyper-box) R package provides two main algorithms: AHB_fast_match and AHB_MIP_match for Interpretable Individualized Treatment Effect Estimation.

alexlanglang.github.io

battleship-game

dame-flame-python-package

A Python Package providing two algorithms, DAME and FLAME, for fast and interpretable treatment-control matches of categorical data

flame_db

FLAME_db is a Python package for performing matching for observational causal inference on datasets containing discrete covariates. It implements the Fast, Large-Scale Almost Matching Exactly (FLAME) algorithms using SQL quries, which match treatment and control units on subsets of the covariates. FLAME_db scales to huge datasets with millions of observations where existing state-of-the-art methods fail, and that it achieves significantly better performance than other matching methods. The resulting matched groups are interpretable, because the matches are made on covariates, and high-quality, because machine learning is used to determine which covariates are important to match on.

improved-regularization-of-convolutional-neural-networks

leaf-damage-semantic-segmentation

We decreased the class imbalance ratio of the leaf dataset to 205 from 415 using LeafImageEditor. The copy/paste data augmentation decreased class imbalance ratio to 50. It helps us to get much better performance on Leaf damage semantic segmentation. The code of algorithm will be published later.

leaf-project

optical-flow-active-learning-release

The official PyTorch implementation of the paper "Optical Flow Training under Limited Label Budget via Active Learning" (ECCV 2022)

pytorch-loss

label-smooth, amsoftmax, focal-loss, triplet-loss, lovasz-softmax. Maybe useful

reram-deep-leanring

We want to find a method to train neural networks with the smallest number of crossbars. Sparse neural networks trained based on The Lottery Ticket Hypothesis enable us to save even 90% of crossbars for training.

risk-game

This is an Android application which enables multiple users to play RISK GAME online concurrently.

wireless-rating-device

This is a wireless scoring board to help blind employees to rate the quality of products in Firmenich Flavor Lab