LydiaXiaohongLi

Albert_Finetune_with_Pretrain_on_Custom_Corpus

1. Pretrain Albert on custom corpus 2. Finetune the pretrained Albert model on downstream task

AllPairShortestPath-Johnson

Implementation of Johnson's algo to calcualte all pair shortest path for a graph in python

AprioriExercise

A simple implementation for frequent pattern mining exercise using Apriori algorithm for a sample data in Python, using Set data structure

DataStreamMedian

Implementation of median extraction given a data stream to continuously extract data stream median using two heap data structure in Python, with implementation of max heap

DijkstraShortestDistance

Implementation of Dijkstra algorithm to calculate shortest distance of a directed graph using Heap with dictionary data structure in Python

KmeansVectorQuantizer

Implementation of vector quantizer using KMeans clustering and test with UCI daily life activity dataset

Megatron-DeepSpeed

Ongoing research training transformer language models at scale, including: BERT & GPT-2

NaiveBayesClassifier

Implementation of Normal and Bernoulli Naive Bayes Classifier, test with pima-indian-diabetes and MNIST dataset and comparison with sklearn Naive Bayes implementation

SentimentClassifierLSTM

Implementation of Bidirectional LSTM with configurable hidden unit, learning rate, dropout and pretrained GloVE word embedding for sentimemt classifier in Keras

TriTraining

Tri Training, Tri Training with Disagreement

LydiaXiaohongLi/Albert_Finetune_with_Pretrain_on_Custom_Corpus

1. Pretrain Albert on custom corpus 2. Finetune the pretrained Albert model on downstream task

LydiaXiaohongLi/TriTraining

Tri Training, Tri Training with Disagreement

LydiaXiaohongLi/Megatron-DeepSpeed

Ongoing research training transformer language models at scale, including: BERT & GPT-2

LydiaXiaohongLi/NaiveBayesClassifier

Implementation of Normal and Bernoulli Naive Bayes Classifier, test with pima-indian-diabetes and MNIST dataset and comparison with sklearn Naive Bayes implementation

LydiaXiaohongLi/KmeansVectorQuantizer

Implementation of vector quantizer using KMeans clustering and test with UCI daily life activity dataset

LydiaXiaohongLi/SentimentClassifierLSTM

Implementation of Bidirectional LSTM with configurable hidden unit, learning rate, dropout and pretrained GloVE word embedding for sentimemt classifier in Keras

LydiaXiaohongLi/AllPairShortestPath-Johnson

Implementation of Johnson's algo to calcualte all pair shortest path for a graph in python

LydiaXiaohongLi/DataStreamMedian

Implementation of median extraction given a data stream to continuously extract data stream median using two heap data structure in Python, with implementation of max heap

LydiaXiaohongLi/DijkstraShortestDistance

Implementation of Dijkstra algorithm to calculate shortest distance of a directed graph using Heap with dictionary data structure in Python

LydiaXiaohongLi/GaussianMixture

Implementation of guassian mixture, with application to image segmentation

LydiaXiaohongLi/HuffmanCoding

Implementation of Huffman coding for a given list of weights using Heap data structure in Python

LydiaXiaohongLi/kernl

Kernl lets you run PyTorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackable.

LydiaXiaohongLi/knapsack

Implementation of Knapsack exact solution by dynamic programming in Python

LydiaXiaohongLi/KruskalCluster

Implementation of Kruskal Clustering algo using Lazy Union Find, union by rank, with path compression in Python

LydiaXiaohongLi/LinearSVMClassifier

Implementation of Linear SVM Classifier with SGD, learning rate decay, test with UC Irvine machine learning Adult Income dataset

LydiaXiaohongLi/map-matching

Map Matching based on GraphHopper

LydiaXiaohongLi/MaxWeightIndependentPath

Implementation of finding Max Weight Independent Path given a path graph using dynamic programming in Python

LydiaXiaohongLi/MeanFieldInferenceDenoiserForBinaryImage

Implementation of mean field infererence for binary boltzmann machine problem to denoise binary image

LydiaXiaohongLi/metricx

LydiaXiaohongLi/MinimumSpanningTree-Prim

Implementation of Prim's algo to find minimum spanning tree using heap data structure in Python

LydiaXiaohongLi/NeMo

NeMo: a toolkit for conversational AI

LydiaXiaohongLi/PCA

Implementation of PCA, test with iris dataset (with additional handcrafted noisy iris data set), compare with sklearn PCA

LydiaXiaohongLi/PoCA

Implementation of PoCA, test and visualize with CIFAR-10 data set, using different distance measures

LydiaXiaohongLi/QuickSorter

Implementation of quick sort for integer arrays in Python

LydiaXiaohongLi/ResNet18

Implementation of ResNet18 and test with CIFAR10

LydiaXiaohongLi/TimeseriesDataCheckClassifierLSTM

Implementation of Bidirectional LSTM with configurable hidden unit, learning rate, dropout for time series data check in Keras.

LydiaXiaohongLi/TravelingSalesmanProb-DynamicProgramming

Implementation of traveling salesman prob exact solution by dynamic programming in python

LydiaXiaohongLi/TravelingSalesmanProb-NearestNeighbour

Implementation of nearest neighbour heuristic for traveling salesman prob in python

LydiaXiaohongLi/VAE

Implementation of variational autoencoder with CNN and generation of fake hand written digit with MNIST dataset

LydiaXiaohongLi/vllm

A high-throughput and memory-efficient inference and serving engine for LLMs