amadeus

アマデウスver 1.0.4

atla-demo

Source code for "Adversarial Training for Layout-Aware Text-VQA".

ats

[ICME 2024] The code for Adversarial Training with OCR Modality Perturbation for Scene-Text Visual Question Answering

attention-efficientzero-alpaca-lora-webui

The Webui based on Alpaca-Lora+ChatGLM aims to visualize Atari game results of Efficientzero.

chatglm-webui

A WebUI for ChatGLM-6B

co-nav-exp

多智能体目标导航实验

cogvlm2-api4

用于softmax分类的CogVLM API

depth_yolo

combination of darknet_ros and iai_kinect2

drawcards

echarts

Apache ECharts is a powerful, interactive charting and data visualization library for browser

efficient-vits-finetuning

Finetuning VITS Efficiently (Lora)

efficientzero

Open-source codebase for EfficientZero, from "Mastering Atari Games with Limited Data" at NeurIPS 2021. Optimize the residual module

frank.github.io

frankzxshen

frankzxshen.github.io

blog，随便创的

gameaudiocrawler

A script used to climb the wiki audios for some common games.

genshin-gacha-export

原神抽卡记录导出

grasscutter

A server software reimplementation for a certain anime game.

habitat-installation

安装habitat的简易流程，以v0.2.2为例

latla

LLM portion of ATLA. Used to bring llama2 external knowledge into Text-VQA.

latr

Implementation of LaTr: Layout-aware transformer for scene-text VQA,a novel multimodal architecture for Scene Text Visual Question Answering (STVQA)

machine-learning-assignments

This project is only for SWJTU's students providing their assignments.

mnlexnet

This is the PyTorch version repository for MNIST dataset identification.

sklearn

from mofan python

so-vits-svc-audio2audio

Replace the song vocals to get the target audio.

tap

TAP: Text-Aware Pre-training for Text-VQA and Text-Caption, CVPR 2021 (Oral)：Add prompt for LLM.

terminal

The new Windows Terminal and the original Windows console host, all in the same place!

visual-chatgpt-zh

visual-chatgpt支持中文版本

visual-chatgpt-zh-vits

visual-chatgpt支持中文的windows版本，融合vits推断模块

vits-fast-fine-tuning

This repo is a pipeline of VITS finetuning for fast speaker adaptation TTS, and many-to-many voice conversion