# 人工智能
# Flag
- https://github.com/luban-agi/Awesome-AIGC-Tutorials (opens new window)
- https://github.com/BradyFU/Awesome-Multimodal-Large-Language-Models (opens new window)
- https://github.com/xx025/carrot (opens new window)
- https://github.com/Charmve/computer-vision-in-action (opens new window)
- https://github.com/alexeygrigorev/mlbookcamp-code (opens new window)
- https://github.com/huggingface (opens new window)
- 辅助驾驶 https://github.com/commaai/openpilot (opens new window)
- https://github.com/OpenDriveLab (opens new window)
- 文档归档 https://github.com/paperless-ngx/paperless-ngx (opens new window)
- https://github.com/xtekky/gpt4free (opens new window)
- https://github.com/JoeSiu/discord-chatbot (opens new window)
- https://github.com/juzeon/poe-openai-proxy (opens new window)
- https://github.com/chathub-dev (opens new window)
- https://github.com/TransformerOptimus/SuperAGI (opens new window)
- https://github.com/openchatai/OpenChat (opens new window)
- https://github.com/sunner/ChatALL (opens new window)
- 语言模型漏洞 https://github.com/llm-attacks/llm-attacks (opens new window)
- ChatGPT/Claude/PaLM/LLaMA/Sage/Dragonfly/Vicuna/Koala/Dolly/ChatGLM/Alpaca 通用型聊天机器人
- https://github.com/pengzhile/pandora (opens new window)
- https://github.com/adams549659584/go-proxy-bingai (opens new window)
- https://github.com/search?q=FreeGPT (opens new window)
- https://github.com/LiLittleCat/awesome-free-chatgpt (opens new window)
- https://github.com/lzwme/chatgpt-sites (opens new window)
- https://github.com/xxxily/hello-ai (opens new window)
- https://chatgptdemo.net (opens new window)
- https://github.com/PlexPt/awesome-chatgpt-prompts-zh (opens new window)
- https://github.com/PromtEngineer/localGPT (opens new window)
- https://github.com/imartinez/privateGPT (opens new window)
- https://github.com/labring/FastGPT (opens new window)
- https://github.com/lm-sys/FastChat (opens new window)
- https://github.com/smol-ai/GodMode (opens new window)
- https://github.com/mylxsw/aidea (opens new window)
- DALL-E 文本生成图片
- Midjourney 文本生成图片
- OpenRead AI读论文
- Jasper.ai 写作AI
- Synthesia 生成真人视频
- Jenni.ai AI写论文
- Copy.ai 写文案
- PepperType A写邮件
- Designs.ai 创意设计平台
- Khroma 高质量的配色方案
- Posemaniacs 人体3D模型
- Aiva 创作音乐
- Al Writer 自动生成新闻稿
- Fireflies 将文本转换为语音
- Repurpose 10 自动化社媒内容
- Chatbot live 多功能聊天机器人
- https://github.com/cocktailpeanut/dalai (opens new window)
- https://github.com/gtoxlili/give-advice (opens new window)
- https://github.com/cocktailpeanut/dalai (opens new window)
- https://github.com/stochasticai/xturing (opens new window)
- https://github.com/facebookresearch/llama (opens new window)
- https://github.com/kennethleungty/Llama-2-Open-Source-LLM-CPU-Inference (opens new window)
- https://github.com/ymcui/Chinese-LLaMA-Alpaca (opens new window)
- https://github.com/rustformers/llama-rs (opens new window)
- https://github.com/karpathy/llama2.c (opens new window)
- https://github.com/LinkSoul-AI/Chinese-Llama-2-7b (opens new window)
- https://github.com/li-plus/chatglm.cpp (opens new window)
- https://github.com/OptimalScale/LMFlow (opens new window)
- 模型调用 https://github.com/hwchase17/langchain (opens new window)
- 金融 https://github.com/AI4Finance-Foundation/FinGPT (opens new window)
# 模型
- https://github.com/topics/machine-learning (opens new window)
- https://github.com/topics/deep-learning (opens new window)
- https://github.com/topics/neural-network (opens new window)
- https://github.com/topics/computer-vision (opens new window)
- https://github.com/topics/artificial-intelligence (opens new window)
- ncnn推理 https://github.com/topics/ncnn (opens new window)
- torch推理 https://github.com/topics/pytorch (opens new window)
- https://github.com/topics/text-recognition (opens new window)
深度学习是神经网络的深层神经网络。强化学习是独立于机器学习之外的分支,强化学习严格来说不属于机器学习, 机器学习的任务是通过已知的训练集来学习目标函数,而强化学习不需要训练集。
- 机器学习:一切通过优化方法挖掘数据中规律的学科。
- 深度学习:一切运用了神经网络作为参数结构进行优化的机器学习算法。学习过程是静态的,训练样本是有标签的
- 强化学习:不仅能利用现有数据,还可以通过对环境的探索获得新数据,并利用新数据循环往复地更新迭代现有模型的机器学习算法。学习过程是动态的,训练是没有标签的
- 深度强化学习:一切运用了神经网络作为参数结构进行优化的强化学习算法。
- 机器学习模型(Machine Learning models):包括分类器、回归器、聚类器、推荐系统等。例如:Logistic回归、深度神经网络、决策树
- Wide & Deep:Google开发的一种混合推荐模型,结合了线性模型(广度)和深度神经网络(深度)。
- Neural Collaborative Filtering (NCF):一种基于深度学习的协同过滤推荐算法。
- DeepFM:一种结合因子分解机(FM)和深度神经网络的推荐算法。
- 强化学习模型(Reinforcement Learning models):使用来自环境的反馈,通过反复试验来学习实现目标的模型。例如:深度Q网络、政策梯度
- 逻辑回归(Logistic Regression)
- K最近邻(K-Nearest Neighbors)
- 决策树(Decision Tree)
- 随机森林(Random Forest)
- 朴素贝叶斯(Naive Bayes)
- 神经网络(Neural Network)
- 感知机(Perceptron)
- 多层感知机(Multilayer Perceptron)
- CNN(Convolutional Neural Network)
- RNN(Recurrent Neural Network)
- LSTM(Long Short Term Memory)
- GAN(Generative Adversarial Nets)
- SVM(Support Vector Machine)
- 聚类(Clustering)
- K均值(K-Means)
- 高斯混合模型(Gaussian Mixture Model)
- DBSCAN
- 降维(Dimensionality Reduction)
- PCA(Principal Component Analysis)
- LDA(Linear Discriminant Analysis)
- -Kernel PCA
- tSNE
- 线性回归(Linear Regression)
- 极限学习机(Extreme Learning Machine)
- 隐马尔可夫模型(Hidden Markov Model)
- 支持向量聚类(Support Vector Clustering)
- 递增学习(Incremental Learning)
- 置信学习(Confidence Learning)
- 元学习(Meta Learning)
- 集成学习(Ensemble Learning)
- 随机森林(Random Forest)
- GBDT(Gradient Boosting Decision Tree)
- AdaBoost
- 栈式学习(Stacking)
- 迁移学习(Transfer Learning)
- 计算机视觉(Computer Vision):用于分析图像和视频的模型。例如:对象检测器、分割模型、面部识别模型
- CNN(Convolutional Neural Network)
- 对象检测(Object Detection)
- semantic segmentation
- 实例分割(Instance Segmentation)
- 目标跟踪(Object Tracking)
- 行人重识别(Person Re-identification)
- 图像分类(Image Classification)
- 物体识别(Object Recognition)
- 自然语言处理(Natural Language Processing):用于处理文本和语音的模型。例如:变形金刚、递归神经网络、单词嵌入
- RNN(Recurrent Neural Network)
- LSTM(Long Short Term Memory)
- GRU(Gated Recurrent Unit)
- 词向量(Word Embedding)
- 文本分类(Text Classification)
- 机器翻译(Machine Translation)
- 问答系统(QA)
- 评论分析(Sentiment Analysis)
- 词性标注(Part-of-speech Tagging)
- 命名实体识别(Named Entity Recognition)
- 依存句法分析(Dependency Parsing)
- 文本摘要(Text Summarization)
- 卷积神经网络(Convolutional Neural Networks,CNN):图像识别、目标检测、图像分割、人脸识别等。
- AlexNet:2012年的ImageNet竞赛冠军,深度卷积神经网络(CNN)的开创者。
- VGG-16 和 VGG-19:牛津大学开发的具有16层和19层的CNN,广泛应用于图像识别任务。
- ResNet (Residual Network):Microsoft开发的一种具有残差连接的深度CNN,解决了深度网络训练中的梯度消失问题。
- Inception 系列(例如Inception v3 和 Inception-ResNet):Google开发的具有多尺度卷积结构的CNN。
- EfficientNet:一种基于自动搜索的CNN,通过调整网络宽度、深度和分辨率实现了高性能和低计算成本的平衡
- 循环神经网络(Recurrent Neural Networks,RNN):自然语言处理、语音识别、时间序列预测等。
- 生成对抗网络(Generative Adversarial Networks,GAN):图像生成、图像超分辨率(Super-Resolution Models)、图像修复、图像风格迁移等。
- SRCNN
- ESPCN
- EDSR
- 转换器(Transformer):自然语言处理、文本分类、机器翻译、问答系统等。
- BERT (Bidirectional Encoder Representations from Transformers):由Google开发的NLP模型,具有双向上下文识别能力。
- GPT (Generative Pre-trained Transformer)系列:由OpenAI开发的一系列模型,包括GPT、GPT-2、GPT-3和GPT-4(即我)。
- T5 (Text-to-Text Transfer Transformer):由Google开发的文本生成和NLP任务的通用模型。
- RoBERTa:Facebook AI的一种优化BERT变体,通过调整训练方法和超参数提高了性能。
- ALBERT (A Lite BERT):一种精简版的BERT,减少了参数数量和计算复杂度。
- XLNet:BERT的双向替代品,结合了Transformer-XL和BERT的优势。
- 预训练语言模型(Pretrained Language Models,PLM):文本分类、情感分析、问答系统、机器翻译等。
- 支持向量机(Support Vector Machine,SVM):分类、回归、异常检测等。
- 决策树(Decision Tree):分类、回归等。
- K均值聚类(K-Means Clustering):聚类分析等。
- YOLO
- Mask R-CNN
- BERT
- GPT
- DQN
# JupyterNotebook
- https://github.com/jupyterlab/jupyterlab (opens new window)
- https://github.com/jupyter (opens new window)
- https://github.com/markusschanta/awesome-jupyter (opens new window)
- jupyter notebook中%与!的区别 (opens new window)
- https://github.com/zjffdu/zeppelin-notebook (opens new window)
# 开源实现
- https://github.com/louisfb01/best_AI_papers_2021 (opens new window)
- https://github.com/pliang279/awesome-multimodal-ml (opens new window)
- https://github.com/roboticcam/machine-learning-notes (opens new window)
- https://github.com/Tencent/ncnn (opens new window)
- https://github.com/onnx/onnx (opens new window)
- https://github.com/ARM-software/ComputeLibrary (opens new window)
- https://github.com/apache/incubator-mxnet (opens new window)
- https://github.com/apache/incubator-systemml (opens new window)
- https://github.com/tensorflow/tensorflow (opens new window)
- https://github.com/Microsoft/CNTK (opens new window)
- https://github.com/deepmind/lab (opens new window)
- https://github.com/pytorch/pytorch (opens new window)
- https://github.com/OpenNMT (opens new window)
- https://github.com/scikit-learn/scikit-learn (opens new window)
- https://github.com/Theano/Theano (opens new window)
- https://github.com/intel/caffe (opens new window)
- https://github.com/BVLC/caffe (opens new window)
- https://github.com/torch/torch7 (opens new window)
- https://github.com/microsoft/lightgbm-transform (opens new window)
- crnn https://github.com/meijieru/crnn.pytorch (opens new window)
- https://github.com/Megvii-BaseDetection/YOLOX (opens new window)
- Psenet https://github.com/WenmuZhou/PSENet.pytorch (opens new window)
- https://github.com/pfnet/pfrl (opens new window)
- 语言模型实现 https://github.com/lukhy/masr (opens new window)
- NLP资源库 https://github.com/fighting41love/funNLP (opens new window)
- https://github.com/ray-project/ray (opens new window)
- 实时语音克隆 https://github.com/CorentinJ/Real-Time-Voice-Cloning (opens new window)
- https://github.com/babysor/Realtime-Voice-Clone-Chinese (opens new window)
- https://github.com/babysor/MockingBird (opens new window)
- https://github.com/cnlinxi/book-text-to-speech (opens new window)
- https://github.com/haoheliu/AudioLDM (opens new window)
- https://github.com/Plachtaa/VALL-E-X (opens new window)
- 音频处理 https://github.com/facebookresearch/audiocraft (opens new window)
- 文本复述 https://github.com/tongchangD/text_data_enhancement_with_LaserTagger (opens new window)
- https://github.com/innnky/so-vits-svc (opens new window)
- https://github.com/34j/so-vits-svc-fork (opens new window)
- https://github.com/suno-ai/bark (opens new window)
- https://github.com/coqui-ai/TTS (opens new window)
- 变声 https://github.com/w-okada/voice-changer (opens new window)
- 捕获键盘声音 https://github.com/ggerganov/kbd-audio (opens new window)
- Whisper https://github.com/chidiwilliams/buzz (opens new window)
- https://github.com/ggerganov/whisper.cpp (opens new window)
- https://github.com/invoke-ai/InvokeAI (opens new window)
- https://github.com/omerbt/TokenFlow (opens new window)
- https://github.com/prompt-engineering (opens new window)
- 视频抠图 https://github.com/PeterL1n/RobustVideoMatting (opens new window)
- https://github.com/senguptaumd/Background-Matting (opens new window)
- 视频制作 https://github.com/elebumm/RedditVideoMakerBot (opens new window)
- 实时屏幕翻译 https://github.com/Danily07/Translumo (opens new window)
- 视频合成 https://github.com/camenduru/text-to-video-synthesis-colab (opens new window)
- 提高分辨率 https://github.com/Janspiry/Image-Super-Resolution-via-Iterative-Refinement (opens new window)
- https://compvis.github.io/taming-transformers (opens new window)
- https://github.com/upscayl/upscayl (opens new window)
- https://github.com/Tohrusky/Final2x (opens new window)
- 图片降噪 https://github.com/google-research/multinerf (opens new window)
- https://github.com/CompVis/latent-diffusion (opens new window)
- https://github.com/AUTOMATIC1111/stable-diffusion-webui (opens new window)
- https://github.com/Nutlope/restorePhotos (opens new window)
- https://github.com/sczhou/CodeFormer (opens new window)
- https://github.com/facefusion/facefusion (opens new window)
- 换脸 https://github.com/iperov/DeepFaceLab (opens new window)
- https://github.com/dunbar12138/pix2pix3D (opens new window)
- https://github.com/phodal/ai-research (opens new window)
- https://github.com/s0md3v/roop (opens new window)
- 换衣服 https://vogue-try-on.github.io (opens new window)
- https://github.com/IDEA-Research/Grounded-Segment-Anything (opens new window)
- https://github.com/modelscope/facechain (opens new window)
- 漫画人脸检测 https://github.com/nagadomi/lbpcascade_animeface (opens new window)
- 卡通化 https://github.com/SystemErrorWang/White-box-Cartoonization (opens new window)
- 图片人物说话 https://github.com/Winfredy/SadTalker (opens new window)
- https://github.com/XingangPan/DragGAN (opens new window)
- 新闻网页正文通用抽取器 https://github.com/GeneralNewsExtractor (opens new window)
- https://github.com/nlplab/brat (opens new window)
- https://github.com/whitead/paper-qa (opens new window)
- 解密加密 https://github.com/Ciphey/Ciphey (opens new window)
- 玩游戏 https://github.com/MineDojo/Voyager (opens new window)
- 提取视频语言 https://github.com/SevaSk/ecoute (opens new window)
- 图像标记 https://github.com/xinyu1205/Recognize_Anything-Tag2Text (opens new window)
- stable-diffusion https://github.com/Fictiverse/Redream (opens new window)
- https://github.com/SupaGruen/StableDiffusion-CheatSheet (opens new window)
- https://github.com/SysCV/sam-hq (opens new window)
- https://github.com/fiatrete/OpenDAN-Personal-AI-OS (opens new window)
- https://github.com/nateraw/stable-diffusion-videos (opens new window)
- https://github.com/Stability-AI/generative-models (opens new window)
- https://github.com/lllyasviel/Fooocus (opens new window)
- https://github.com/rese1f/StableVideo (opens new window)
- https://github.com/CloudOrc/SolidUI (opens new window)
- https://github.com/ainize-team (opens new window)
- 动手学深度学习 https://github.com/d2l-ai (opens new window)
- 视频感知 https://github.com/open-mmlab/mmtracking (opens new window)
- https://github.com/scikit-learn/scikit-learn (opens new window)
- 数据管道 https://github.com/orchest (opens new window)
- 数据科学Web https://github.com/plotly/dash (opens new window)
- 扫描PDF https://github.com/baicunko/scanyourpdf (opens new window)
- https://github.com/eosphoros-ai/DB-GPT (opens new window)
- 签名 https://github.com/SigmaHQ/sigma (opens new window)
- https://github.com/Alic-yuan/nlp-beginner-finish (opens new window)
- https://github.com/heartexlabs/label-studio (opens new window)
- https://github.com/YunwenTechnology/Unilm (opens new window)
- 手机智能机器人 https://github.com/isl-org/OpenBot (opens new window)
- https://github.com/imistyrain/MTCNN (opens new window)
- 医疗 https://github.com/DengYangyong/medical_entity_recognize (opens new window)
- 摄像头物体识别 https://github.com/blakeblackshear/frigate (opens new window)
# OCR
- https://github.com/topics/ocr (opens new window)
- https://github.com/topics/ocr-recognition (opens new window)
- https://github.com/search?q=ocr (opens new window)
ImportError: libGL.so.1: cannot open shared object file: No such file or directory
apt install libgl1-mesa-glx
debianyum install mesa-libGL-devel
centos
Could not find library geos_c or load any of its variants ['libgeos_c.so.1', 'libgeos_c.so']
apt install libgeos-dev
debianyum install geos-devel
centos
- https://github.com/PearOCR (opens new window)
- https://github.com/alisen39/TrWebOCR (opens new window)
pip3 install libtorch numpy opencv-python Pillow tornado
- https://github.com/ianzhao05/textshot (opens new window)
- https://github.com/ouyanghuiyu/chineseocr_lite (opens new window)
pip3 install tornado numpy opencv_python onnxruntime Shapely pyclipper Pillow
- https://github.com/chineseocr (opens new window)
- https://github.com/facebookresearch/nougat (opens new window)
- https://github.com/Greedysky/TTKOCR (opens new window)
- https://github.com/tesseract-ocr/tesseract (opens new window)
- 百度飞桨 https://github.com/PaddlePaddle/PaddleOCR (opens new window)
- https://github.com/miaomiaosoft/PandaOCR (opens new window)
- https://github.com/WenmuZhou/PytorchOCR (opens new window)
- https://github.com/microsoft/unilm (opens new window)
- 语音识别 github.com/microsoft/NeuralSpeech