leocnj

keras_attn_reader

Keras implementation of an attention reader (using LSTM or GRU RNN)

alpaca-lora

Instruct-tuning LLaMA on consumer hardware

asap

ASAP: A Chinese Review Dataset Towards Aspect Category Sentiment Analysis and Rating Prediction

attention

some attention implements

call2018

cv_flask

flask server running CV tasks.

deep-learning-conda-envs

This repo contains Anaconda environment.yml files for different deep learning frameworks (i.e. TensorFlow, PyTorch, Theano, etc.)

DeepSpeedExamples

Example models using DeepSpeed

dlib_try

Try dlib, cv2 funcitons.

docker-images

Dockerfiles for some multimodal toolkits

docker-openpose

docker-pyaa

Dockerfile for pyAudioAnalysis

ecom-rakuten

Winning solution for the Rakuten Data Challenge, as part of SIGIR eCom '18.

face-classification-js

Face classification on JavaScript

ichiba_imgs

kaldi-dnn-ali-gop

Forced alignment and Goodness of Pronunciation (GOP) with DNN support. Bases on Kaldi.

KOBE

Source code and dataset for KDD 2019 paper "Towards Knowledge-Based Personalized Product Description Generation in E-commerce"

leocnj.github.io

lm-evaluation-harness

A framework for few-shot evaluation of language models.

LongLoRA

Efficient long-context fine-tuning, supervised fine-tuning, LongQA dataset.

Megatron-LM

Ongoing research training transformer models at scale

MixMatch-pytorch

Code for "MixMatch - A Holistic Approach to Semi-Supervised Learning"

nanotron

Minimalistic large language model 3D-parallelism training

pytorch-kaldi-old

pytorch-kaldi is a project for developing state-of-the-art DNN/RNN hybrid speech recognition systems. The DNN part is managed by pytorch, while feature extraction, label computation, and decoding are performed with the kaldi toolkit.

pytorch-StarGAN-VC

Fully reproduce the paper of StarGAN-VC. Stable training and Better audio quality .

RAG_learn

A learning project of RAG and LLM

StarGAN-Voice-Conversion

This is a pytorch implementation of the paper: StarGAN-VC: Non-parallel many-to-many voice conversion with star generative adversarial networks

Video-Expression-Recognition

Recognising expression/emotion of unique faces in a video

video_proc_docker

run openface (in a docker) on a video dir

vllm

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