├── ComputerVision.md ├── DataScience.md ├── LICENSE ├── NLP.md ├── NLP ├── Chatbot_Evaluation_Metrics.md └── Practical_text_to_SQL_for_data_analytics_by_linkedin.md ├── README.md └── Speech.md /ComputerVision.md: -------------------------------------------------------------------------------- 1 | # Great Deep Learning Tutorials for Compter Vision 2 | A Great Collection of Deep Learning Tutorials and Repositories for Compter Vision 3 | 4 | ## General: 5 | - [Microsoft Computer Vision Recipes](https://github.com/microsoft/computervision-recipes) [_Great_] 6 | - [OpenMMLab](https://openmmlab.com/) [_Great_] 7 | - [OpenMMLab - GitHub](https://github.com/open-mmlab) 8 | - [OpenMMLab - MMCV is a foundational python library for computer vision research](https://github.com/open-mmlab/mmcv) 9 | - [OpenMMLab - MMEditing is an open source image and video editing toolbox](https://github.com/open-mmlab/mmediting) 10 | - [OpenMMLab - MMDetection](https://github.com/open-mmlab/mmdetection) 11 | - [Kornia - is a differentiable computer vision library for PyTorch](https://github.com/kornia/kornia) 12 | - [Great Computer Vision Tutorials and Notebooks](https://github.com/roboflow/notebooks) 13 | 14 | ## Model Visualization: 15 | - [CNN Visualizations](https://github.com/utkuozbulak/pytorch-cnn-visualizations) [_Very Good_] 16 | - [CNN-heatmap](https://github.com/mrgloom/CNN-heatmap) 17 | - [Tools to Design or Visualize Architecture of Neural Network](https://github.com/ashishpatel26/Tools-to-Design-or-Visualize-Architecture-of-Neural-Network) [**Great**] 18 | - [Netron - GitHub](https://github.com/lutzroeder/netron) [_Excellent_] 19 | - [Monitor your GPUs](https://github.com/msalvaris/gpu_monitor) [**Excellent**] 20 | - [Understanding CNN](http://cs231n.github.io/understanding-cnn/) 21 | - [Exploring Neural Networks with Activation Atlases](https://distill.pub/2019/activation-atlas/) [_Great_] 22 | - [Explaining What Explainable AI Did Not](https://bair.berkeley.edu/blog/2020/04/23/decisions/) [_Interesting_] 23 | - [CNN Explainer](https://poloclub.github.io/cnn-explainer/) [_Interesting_] 24 | - [Interactive Tools for ML, DL and Math](https://github.com/Machine-Learning-Tokyo/Interactive_Tools#embedding-projector) [_Interesting_] 25 | - [Visualizing Neural Networks with the Grand Tour](https://distill.pub/2020/grand-tour/) [_Interesting_] 26 | - [Zoom In: An Introduction to Circuits](https://distill.pub/2020/circuits/zoom-in/) 27 | - [Concept: Concept Modeling on Images](https://github.com/MaartenGr/Concept) 28 | 29 | ### GradCAM: 30 | - [GradCAM Main Paper](https://arxiv.org/abs/1610.02391) 31 | - [PyTorch GradCAM 1](https://github.com/jacobgil/pytorch-grad-cam) 32 | - [PyTorch GradCAM 2](https://github.com/kazuto1011/grad-cam-pytorch) 33 | - [Keras GradCAM](https://keras.io/examples/vision/grad_cam/) 34 | - [pyimagesearch GradCAM](https://www.pyimagesearch.com/2020/03/09/grad-cam-visualize-class-activation-maps-with-keras-tensorflow-and-deep-learning/) 35 | 36 | ### t-SNE Visualization: 37 | - [TSNE-Visualization of large dataset images using pre-trained networks in Tensorflow and Keras](https://e3oroush.github.io/tsne-visualization/) [_Great_] 38 | 39 | ### UMAP Visualization: 40 | - [Understanding UMAP](https://pair-code.github.io/understanding-umap/) 41 | 42 | ### EDA and visualization of Image/Video Datasets: 43 | - [fastdup: a powerful free tool designed to rapidly extract valuable insights from your image & video datasets](https://github.com/visual-layer/fastdup) 44 | 45 | ## Image Classification Models: 46 | - [GhostNet (CVPR 2020) in PyTorch and TensorFlow](https://blog.paperspace.com/ghostnet-cvpr-2020/) 47 | - [GhostNet - GitHub](https://github.com/huawei-noah/ghostnet) 48 | - [GhostNet - PyTorch Hub](https://pytorch.org/hub/pytorch_vision_ghostnet/) [_Excellent_] 49 | - [Residual blocks — Building blocks of ResNet](https://towardsdatascience.com/residual-blocks-building-blocks-of-resnet-fd90ca15d6ec) 50 | - [EfficientNet-PyTorch](https://github.com/lukemelas/EfficientNet-PyTorch) 51 | - [EfficientNet Explanation](https://amaarora.github.io/2020/08/13/efficientnet.html) 52 | - [DeepMind - NFNets](https://github.com/deepmind/deepmind-research/tree/master/nfnets) 53 | - [NFNets - PyTorch](https://github.com/vballoli/nfnets-pytorch) 54 | - [EfficientNetV2](https://arxiv.org/pdf/2104.00298.pdf) 55 | - [Deit: Data-Efficient architectures and training for Image classification](https://github.com/facebookresearch/deit) 56 | - [How to Train State-Of-The-Art Models Using TorchVision’s Latest Primitives](https://pytorch.org/blog/how-to-train-state-of-the-art-models-using-torchvision-latest-primitives/) [_Excellent_] 57 | 58 | ### Data Augmentation for Image Classification: 59 | - [Image Augmentation](https://github.com/aleju/imgaug) 60 | - [AugLy - data augmentations library that supports different modalities](https://github.com/facebookresearch/AugLy) 61 | - [Learnable Test-time Augmentation](https://github.com/bayesgroup/gps-augment) 62 | 63 | ## Self-Supervised Learning: 64 | - [VISSL - VIsion library for state-of-the-art Self-Supervised Learning](https://vissl.ai/) 65 | - [VISSL - GitHub](https://github.com/facebookresearch/vissl) 66 | - [DINO: Self-Supervised Vision Transformers](https://github.com/facebookresearch/dino) 67 | - [DINOv2: Learning Robust Visual Features without Supervision](https://github.com/facebookresearch/dinov2) 68 | - [I-JEPA (the Image-based Joint-Embedding Predictive Architecture)](https://github.com/facebookresearch/ijepa) 69 | 70 | ## Transformers in Computer Vision: 71 | - [Vision Transformers Tutorial](https://blog.paperspace.com/vision-transformers/) [**Great**] 72 | - [Transformers in computer vision: ViT architectures, tips, tricks and improvements](https://theaisummer.com/transformers-computer-vision/) [Great] 73 | - [Vision Transformer](https://github.com/google-research/vision_transformer) 74 | - [Vision Transformer - Pytorch](https://github.com/lucidrains/vit-pytorch) [**Great**] 75 | - [PyTorch External Attention](https://github.com/xmu-xiaoma666/External-Attention-pytorch) [**Great**] 76 | - [MobileViT in PyTorch](https://github.com/chinhsuanwu/mobilevit-pytorch) 77 | - [Clip-vit](https://huggingface.co/openai/clip-vit) 78 | - [OpenAI CLIP](https://github.com/openai/CLIP) 79 | - [Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model](https://www.arxiv.org/abs/2408.11039) 80 | - [CLOC: Contrastive Localized Language-Image Pre-Training](https://www.linkedin.com/posts/zhe-gan-a2229a78_clip-is-the-default-choice-for-most-multimodal-activity-7248009680340320256--dqE?utm_source=share&utm_medium=member_desktop) 81 | - [Transformers: from NLP to CV](https://github.com/IbrahimSobh/Transformers) [**Very Great & Excellent**] 82 | 83 | ### Vision-Language Models (LLMs in Image & Computer Vision): 84 | - [Prismer: A Vision-Language Model](https://github.com/NVlabs/prismer) 85 | - [ViperGPT: Visual Inference via Python Execution for Reasoning](https://github.com/cvlab-columbia/viper) 86 | - [LLaVA: Large Language and Vision Assistant](https://llava-vl.github.io/) 87 | - [ImageBind-LLM](https://github.com/OpenGVLab/LLaMA-Adapter/tree/main/imagebind_LLM) 88 | - [An Introduction to Vision-Language Modeling](https://www.linkedin.com/posts/aiatmeta_new-from-fair-an-introduction-to-vision-language-activity-7201264894791335938-8LnO?utm_source=share&utm_medium=member_desktop) 89 | - [A comprehensive tutorial on building Vision-Language Models (VLMs)](https://www.linkedin.com/posts/andresmarafioti_excited-to-share-that-our-idefics-3-and-activity-7234088501271502848-D-aY?utm_source=share&utm_medium=member_desktop) 90 | - [Video Web Arena: agent models for OS and web control with memory](https://www.linkedin.com/posts/rogeriobonatti_new-agent-benchmarks-for-os-and-website-control-activity-7257095266980237315-nyqT?utm_source=share&utm_medium=member_desktop) 91 | - [Qwen2-VL: To See the World More Clearly](https://qwenlm.github.io/blog/qwen2-vl/) 92 | 93 | ### Multi-Modal LLMs: 94 | - [CoDi: Any-to-Any Generation via Composable Diffusion](https://github.com/microsoft/i-Code/tree/main/i-Code-V3) 95 | - [kosmos-2: Grounding Multimodal Large Language Models to the World](https://github.com/microsoft/unilm/tree/master/kosmos-2) 96 | 97 | ### Language-Vision Intelligence: 98 | - [Salesforce LAVIS: A Library for Language-Vision Intelligence](https://github.com/salesforce/LAVIS) 99 | 100 | ## Generative AI in Image and Computer Vision: 101 | - [Dream Gaussian: Generative Gaussian Splatting for Efficient 3D Content Creation](https://github.com/dreamgaussian/dreamgaussian) 102 | - [Stable Diffusion Video](https://www.linkedin.com/posts/eric-vyacheslav-156273169_stability-just-released-stable-diffusion-activity-7133533117700521984-JWpc?utm_source=share&utm_medium=member_desktop) 103 | - [Tutorial on Diffusion Models for Imaging and Vision](https://arxiv.org/abs/2403.18103) 104 | 105 | ## Detection & Segmentation: 106 | ## Object Detection: 107 | - [Object Detection for Dummies Part 1](https://lilianweng.github.io/lil-log/2017/10/29/object-recognition-for-dummies-part-1.html) 108 | - [Object Detection for Dummies Part 2](https://lilianweng.github.io/lil-log/2017/12/15/object-recognition-for-dummies-part-2.html) 109 | - [Object Detection for Dummies Part 3](https://lilianweng.github.io/lil-log/2017/12/31/object-recognition-for-dummies-part-3.html) 110 | - [Object Detection for Dummies Part 4](https://lilianweng.github.io/lil-log/2018/12/27/object-detection-part-4.html) 111 | - [Open MMLab Detection Toolbox and Benchmark](https://github.com/open-mmlab/mmdetection) 112 | - [RetinaNet: how Focal Loss fixes Single-Shot Detection](https://towardsdatascience.com/retinanet-how-focal-loss-fixes-single-shot-detection-cb320e3bb0de) 113 | - [Getting Started With Bounding Box Regression In TensorFlow](https://towardsdatascience.com/getting-started-with-bounding-box-regression-in-tensorflow-743e22d0ccb3) 114 | - [Pelee: A Real-Time Object Detection System on Mobile Devices](https://github.com/Robert-JunWang/Pelee) 115 | - [Pelee: Tutorial](https://medium.com/@bibekchaudhary/pelee-real-time-object-detection-system-on-mobile-devices-f565947c04c4) 116 | - [An overview of deep-learning based object-detection algorithms](https://medium.com/@fractaldle/brief-overview-on-object-detection-algorithms-ec516929be93) 117 | - [Object detection and tracking in PyTorch](https://towardsdatascience.com/object-detection-and-tracking-in-pytorch-b3cf1a696a98) 118 | - [Object Detection with RetinaNet - Keras](https://keras.io/examples/vision/retinanet/) 119 | - [PP-YOLO](https://github.com/PaddlePaddle/PaddleDetection/blob/release/0.4/configs/ppyolo/README.md) 120 | - [Super Fast and Accurate 3D Object Detection based on 3D LiDAR Point Clouds](https://github.com/maudzung/Super-Fast-Accurate-3D-Object-Detection) 121 | - [Signfeld - Synthetic traffic sign detection](https://github.com/moabitcoin/signfeld) 122 | - [Swin-Transformer](https://github.com/microsoft/Swin-Transformer) 123 | - [DETR - DEtection TRansformer](https://github.com/facebookresearch/detr) [_Great_] 124 | - [YOLOX](https://github.com/Megvii-BaseDetection/YOLOX) 125 | - [Yolo-v5 vs EfficientDet](https://blog.roboflow.com/yolov5-is-here/) 126 | - [MMRotate: OpenMMLab Rotated Object Detection](https://github.com/open-mmlab/mmrotate) [**Great**] 127 | - [YOLO v7](https://github.com/WongKinYiu/yolov7) 128 | - [Ultralytics YOLOv8](https://github.com/ultralytics/ultralytics) 129 | - [Yolov8 Tracking](https://github.com/mikel-brostrom/yolov8_tracking) 130 | - [YOLO-NAS](https://github.com/Deci-AI/super-gradients/blob/master/YOLONAS.md) 131 | - [Object Detection Leaderboard](https://roboflow.github.io/model-leaderboard/) 132 | - [Object Detection Leaderboard Linkedin Post](https://www.linkedin.com/posts/skalskip92_computervision-objectdetection-opensource-activity-7239562459794096128-E6KR?utm_source=share&utm_medium=member_desktop) 133 | 134 | ### Augmentation for Object Detection & Instance Segmentation: 135 | - [Albumentations - Great Library for image augmentation and transformations](https://github.com/albumentations-team/albumentations) 136 | - [Shapely - Manipulation and analysis of geometric objects](https://github.com/Toblerity/Shapely) 137 | 138 | ### Faster R-CNN Object Detector Tutorials: 139 | - [Guide to build Faster RCNN in PyTorch](https://medium.com/@fractaldle/guide-to-build-faster-rcnn-in-pytorch-95b10c273439) 140 | - [Simple and Fast Implementation of Faster R-CNN](https://github.com/chenyuntc/simple-faster-rcnn-pytorch) 141 | - [PyTorch Tutorial to Object Detection](https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Object-Detection) 142 | - [Object Detection and Classification using R-CNNs](http://www.telesens.co/2018/03/11/object-detection-and-classification-using-r-cnns/) 143 | - [Faster R-CNN: Down the rabbit hole of modern object detection](https://tryolabs.com/blog/2018/01/18/faster-r-cnn-down-the-rabbit-hole-of-modern-object-detection/) 144 | - [Faster R-CNN (object detection) implemented by Keras for custom data](https://towardsdatascience.com/faster-r-cnn-object-detection-implemented-by-keras-for-custom-data-from-googles-open-images-125f62b9141a) 145 | - [Mask R-CNN Unmasked](https://medium.com/@fractaldle/mask-r-cnn-unmasked-c029aa2f1296) 146 | 147 | ### Instance Segmentation: 148 | - [Image Segmentation: tips and tricks from 39 Kaggle competitions](https://neptune.ai/blog/image-segmentation-tips-and-tricks-from-kaggle-competitions?utm_source=reddit&utm_medium=post&utm_campaign=blog-image-segmentation-tips-and-tricks-from-kaggle-competitions) [_Excellent_] 149 | 150 | ### Semantic Segmentation: 151 | - [Simple Guide to Semantic Segmentation](https://medium.com/beyondminds/a-simple-guide-to-semantic-segmentation-effcf83e7e54) 152 | - [Background Matting](https://github.com/PeterL1n/BackgroundMattingV2) 153 | - [Facebook's Segment Anything Model (SAM)](https://github.com/facebookresearch/segment-anything) 154 | - [SAM 2 by Meta: Segment Anything in Images and Videos](https://github.com/facebookresearch/segment-anything-2) 155 | 156 | ### License Plate Detection and Recognition: 157 | - [ALPR in Unscontrained Scenarios](https://github.com/sergiomsilva/alpr-unconstrained) [_Good_] 158 | - [ALPR in Unscontrained Scenarios - Project Page](http://sergiomsilva.com/pubs/alpr-unconstrained/) 159 | - [Lisence plate detection and recognition - ANPR](https://github.com/GuiltyNeuron/ANPR) 160 | - [License Plate Detection](https://github.com/AzharMithani/License-Plate-Detection) 161 | - [yolo3 to detection license plate](https://github.com/ThorPham/License-plate-detection) 162 | 163 | ### Keypoint Regression: 164 | - [Transfer Learning Library](https://github.com/thuml/Transfer-Learning-Library) 165 | - [DEKR: dense keypoint regression framework](https://github.com/HRNet/DEKR) 166 | - [CenterNet](https://github.com/xingyizhou/CenterNet) 167 | - [AdelaiDet](https://github.com/aim-uofa/AdelaiDet) 168 | - [Keypoint Regression](https://github.com/Srijan2001/Keypoint-regression) 169 | 170 | ## Loss Functions: 171 | - [Use Focal Loss To Train Model Using Imbalanced Dataset](https://leimao.github.io/blog/Focal-Loss-Explained/) 172 | 173 | ## Image Augmentation: 174 | - [Image Augmentation for more train data - Kaggle](https://www.kaggle.com/vishnurapps/image-augmentation-for-more-training-data) [_Great_] 175 | 176 | ## Deformable Convolution PyTorch Implementation: 177 | - [TorchVision - Deformable Convolution - Link1](https://github.com/pytorch/vision/releases) 178 | - [TorchVision - Deformable Convolution - Link2](https://github.com/pytorch/vision/blob/master/torchvision/ops/deform_conv.py) 179 | - [Simple PyTorch Deformable Convolution v2](https://github.com/developer0hye/Simple-PyTorch-Deformable-Convolution-v2) [Great] 180 | - [Deformable Convolutional Networks v2 with Pytorch](https://github.com/CharlesShang/DCNv2/tree/pytorch_1.0) [_Good_] 181 | - [Deformable ConvNets v2](https://github.com/4uiiurz1/pytorch-deform-conv-v2) 182 | - [MMLab Detection Toolbox](https://github.com/open-mmlab/mmdetection/tree/master/mmdet/ops/dcn) 183 | - [DCNv2 in PyTorch](https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch) 184 | - [PyTorch implementation of Deformable Convolution](https://github.com/oeway/pytorch-deform-conv) 185 | 186 | 187 | ## 3D Image Classification & 3D Computer Vision: 188 | - [3D MNIST](https://medium.com/shashwats-blog/3d-mnist-b922a3d07334) 189 | - [3D CNN Images with Tensorflow](https://github.com/jibikbam/CNN-3D-images-Tensorflow) 190 | - [Point Cloud Library (PCL)](https://github.com/PointCloudLibrary/pcl) 191 | - [Kaolin - a PyTorch library for accelerating 3D deep learning](https://github.com/NVIDIAGameWorks/kaolin) 192 | - [PyTorch3D](https://github.com/facebookresearch/pytorch3d) [_Fantastic_] 193 | - [OpenCV Tutorial Homography](https://docs.opencv.org/master/d9/dab/tutorial_homography.html) 194 | - [Total 3D Understanding](https://github.com/yinyunie/Total3DUnderstanding) 195 | 196 | ## OCR: 197 | - [Easy OCR](https://github.com/JaidedAI/EasyOCR) [**Great**] 198 | - [MMOCR](https://github.com/open-mmlab/mmocr) [**Great**] 199 | - [Word Level OCR Dataset for Persian Language](https://github.com/amirabbasasadi/Shotor) 200 | - [Simple Persian Word-Level OCR](https://github.com/amirabbasasadi/PersianOCR) 201 | - [Amazon Textract - OCR](https://aws.amazon.com/textract/) 202 | - [TextFuseNet](https://github.com/ying09/TextFuseNet) 203 | - [Transformer-OCR](https://github.com/fengxinjie/Transformer-OCR) 204 | - [Microsoft TrOCR](https://github.com/microsoft/unilm/tree/master/trocr) 205 | - [vedastr: open source scene text recognition toolbox](https://github.com/Media-Smart/vedastr) 206 | - [Goodbye OCR - Welcome Donut from MIT](https://www.linkedin.com/posts/aagarwal29_sagemaker-donut-mit-activity-7067002244352651265-yvfo/?utm_source=share&utm_medium=member_android) 207 | - [Nougat: Neural Optical Understanding for Academic Documents](https://github.com/facebookresearch/nougat) 208 | - [Accurate line-level text detection and recognition (OCR) in any language](https://github.com/VikParuchuri/surya) 209 | 210 | ### Persian OCR: 211 | - [Persian Dataset](https://github.com/persiandataset) 212 | - [Arshasb: Persian OCR dataset](https://github.com/persiandataset/Arshasb) 213 | - [Above Data Set (complete set)](https://drive.google.com/file/d/1Ahw_y-ugnu6FFz_SRBm2VUxnaqmXfsvp) 214 | 215 | ## Few-Shot Learning: 216 | - [Advances in few-shot learning](https://towardsdatascience.com/advances-in-few-shot-learning-a-guided-tour-36bc10a68b77) 217 | - [Advances in few-shot learning: reproducing results in PyTorch](https://towardsdatascience.com/advances-in-few-shot-learning-reproducing-results-in-pytorch-aba70dee541d) 218 | - [One Shot learning, Siamese networks and Triplet Loss](https://medium.com/@crimy/one-shot-learning-siamese-networks-and-triplet-loss-with-keras-2885ed022352) 219 | - [Meta-Learning with Differentiable Convex Optimization](https://github.com/kjunelee/MetaOptNet) 220 | - [Building a One-Shot Learning Network with PyTorch](https://towardsdatascience.com/building-a-one-shot-learning-network-with-pytorch-d1c3a5fafa4a) 221 | - [Annotation-Efficient Learning](https://annotation-efficient-learning.github.io/) [_Good Few-Shot Learning Tutorial_] 222 | - [Awesome Papers - Few shot](https://github.com/Duan-JM/awesome-papers-fewshot) 223 | - [Few-Shot-Learning](https://github.com/daooshee/Few-Shot-Learning) 224 | - [OpenMMLab FewShot Learning](https://github.com/open-mmlab/mmfewshot) [Great] 225 | 226 | ## Learning to Hash & General Hashing (+ Fast Searching Methods): 227 | - [Finding similar images using Deep learning and Locality Sensitive Hashing](https://towardsdatascience.com/finding-similar-images-using-deep-learning-and-locality-sensitive-hashing-9528afee02f5) [_Very Good_] 228 | - [Image similarity using Triplet Loss](https://towardsdatascience.com/image-similarity-using-triplet-loss-3744c0f67973) 229 | - [Finding duplicate images made easy!](https://github.com/idealo/imagededup) 230 | - [Duplicate Image Detection - perspective hash (pHash)](https://github.com/philipbl/duplicate-images) 231 | - [ImageHash](https://github.com/JohannesBuchner/imagehash) 232 | - [NGT - Neighborhood Graph and Tree for Indexing High-dimensional Data](https://github.com/yahoojapan/NGT) [_Great_] 233 | - [NGT - Tutorial](https://github.com/yahoojapan/NGT/wiki) 234 | - [NGT - Python](https://github.com/yahoojapan/NGT/blob/master/python/README.md) 235 | - [Visual Search with MXNet Gluon and HNSW](https://github.com/ThomasDelteil/VisualSearch_MXNet) 236 | - [Annoy: Approximate Nearest Neighbors in C++/Python](https://github.com/spotify/annoy) [**Great**] 237 | - [datasketch: Big Data Looks Small](https://github.com/ekzhu/datasketch) [**Great:** probabilistic data structures that can process and search very large amount of data super fast] 238 | - [Image Similarity with Hugging Face Datasets and Transformers](https://huggingface.co/blog/image-similarity) 239 | 240 | ## Video Understanding: 241 | - [Holistic Video Understanding Challenge](https://holistic-video-understanding.github.io/workshops/iccv2019.html) 242 | - [Holistic Video Understanding Dataset](https://github.com/holistic-video-understanding/Mini-HVU) 243 | - [Holistic Large Scale Video Understanding - Tutorial](https://towardsdatascience.com/holistic-large-scale-video-understanding-c423701b777a) 244 | - [Deep Learning on Video - Part1](https://towardsdatascience.com/deep-learning-on-video-part-one-the-early-days-8a3632ed47d4) 245 | - [VMZ: Model Zoo for Video Modeling](https://github.com/facebookresearch/VMZ) 246 | - [The 3rd YouTube-8M Video Understanding Challenge - 1st Place Solution](https://www.kaggle.com/c/youtube8m-2019/discussion/112869#latest-668179?utm_medium=email&utm_source=intercom&utm_campaign=competition-recaps-yt-2019) 247 | - [torchvideo](https://torchvideo.readthedocs.io/en/latest/#) 248 | - [torchvideo - GitHub](https://github.com/torchvideo/torchvideo) 249 | - [PytorchVideo - GitHub](https://github.com/facebookresearch/pytorchvideo) [_Great_] 250 | - [PytorchVideo - main page](https://pytorchvideo.org/) [_Great_] 251 | - [MoViNet-pytorch](https://github.com/Atze00/MoViNet-pytorch) [Interesting] 252 | - [Basic Video transforms for PyTorch](https://github.com/hassony2/torch_videovision) 253 | - [Fast and Easy to use video feature extractor](https://github.com/antoine77340/video_feature_extractor) 254 | - [Video Augmentation Techniques for Deep Learning](https://github.com/okankop/vidaug) [_Great_] 255 | - [Decord_loader - Excellent Video Data Loader](https://gluon-cv.mxnet.io/build/examples_action_recognition/decord_loader.html) [_Great_] 256 | - [Decord - GitHub](https://github.com/dmlc/decord) 257 | - [PyAV - Pythonic binding for the FFmpeg libraries](https://github.com/PyAV-Org/PyAV) 258 | - [Python bindings for FFmpeg](https://github.com/kkroening/ffmpeg-python) 259 | - [GStreamer: Multimedia Framework](https://github.com/GStreamer/gstreamer) 260 | - [NÜWA: text to video synthesis](https://github.com/lucidrains/nuwa-pytorch) 261 | - [Awesome Video Datasets](https://github.com/xiaobai1217/Awesome-Video-Datasets) 262 | - [Iranian Movies Kaggle Dataset](https://www.kaggle.com/datasets/arianghasemi/iranian-movies) 263 | - [Kinetics 400 Data Set - Download Link via DropBox](https://www.dropbox.com/s/wcs01mlqdgtq4gn/compress.tar.gz?dl=1) 264 | 265 | 266 | ## Text-to-Video: 267 | - [Text to Video Synthesis](https://huggingface.co/spaces/damo-vilab/modelscope-text-to-video-synthesis) 268 | 269 | ## Optical Flow: 270 | - [RAFT](https://github.com/princeton-vl/RAFT) 271 | 272 | 273 | ## Visual Attention Method: 274 | - [Attention Mechanisms in Computer Vision Part 1: CBAM](https://blog.paperspace.com/attention-mechanisms-in-computer-vision-cbam/) [_Excellent_] 275 | - [Efficient Channel Attention for Deep Convolutional Neural Networks (ECA-Net)](https://blog.paperspace.com/attention-mechanisms-in-computer-vision-ecanet/) 276 | - [ECA-Net: Efficient Channel Attention](https://github.com/BangguWu/ECANet) 277 | - [Channel Attention and Squeeze-and-Excitation Networks (SENet)](https://blog.paperspace.com/channel-attention-squeeze-and-excitation-networks/) 278 | - [Self-Attention In Computer Vision](https://towardsdatascience.com/self-attention-in-computer-vision-2782727021f6) 279 | 280 | 281 | ## Visual Question Answering (VQA): 282 | - [Vanilla VQA](https://medium.com/ai2-blog/vanilla-vqa-adcaaaa94336) 283 | 284 | 285 | ## Pose Estimation: 286 | - [Pose Animator](https://github.com/yemount/pose-animator/) 287 | - [RepNet: Weakly Supervised 3D Human Pose Estimation](https://github.com/bastianwandt/RepNet) 288 | 289 | 290 | ## Object Tracking: 291 | - [GSDT](https://github.com/yongxinw/GSDT) 292 | 293 | 294 | ## GANs: 295 | - [MMGeneration](https://github.com/open-mmlab/mmgeneration) 296 | - [Generate Anime Style Face Using DCGAN and Explore Its Latent Feature Representation](https://towardsdatascience.com/generate-anime-style-face-using-dcgan-and-explore-its-latent-feature-representation-ae0e905f3974) 297 | - [StyleGAN2](https://github.com/EvgenyKashin/stylegan2-distillation) 298 | - [DeepFaceLab](https://github.com/iperov/DeepFaceLab) 299 | - [imagen-pytorch: Google's Text-to-Image Neural Network](https://github.com/lucidrains/imagen-pytorch) [Great] 300 | - [DragGAN](https://github.com/XingangPan/DragGAN) 301 | 302 | ## Image Super-Resolution: 303 | - [Image Super-Resolution: A Comprehensive Review](https://blog.paperspace.com/image-super-resolution/) 304 | 305 | 306 | ## Siamese Networks: 307 | - [Siamese Network Tutorial with TensorFlow](https://leimao.github.io/article/Siamese-Network-MNIST/) 308 | 309 | 310 | ## Model Evaluation: 311 | - [Precision and Recall for Multi-Class](https://stats.stackexchange.com/questions/51296/how-do-you-calculate-precision-and-recall-for-multiclass-classification-using-co) 312 | - [Precision and Recall](https://en.wikipedia.org/wiki/Precision_and_recall) 313 | - [Precision-Recall metric](https://scikit-learn.org/stable/auto_examples/model_selection/plot_precision_recall.html) 314 | 315 | 316 | ## Model Size & Speed of Models: 317 | - [GFLOPs & Number of Parameters of ResNet Models](https://neurohive.io/wp-content/uploads/2019/01/resnet-architectures-34-101.png) 318 | - [EfficientNet Models Accuracy vs FLOPs](https://raw.githubusercontent.com/tensorflow/tpu/master/models/official/efficientnet/g3doc/flops.png) 319 | 320 | 321 | ## Computer Vision Annotation Tools: 322 | - [CVAT](https://github.com/opencv/cvat) 323 | - [SuperAnnotate](https://www.superannotate.com/) 324 | - [OpenCV - SuperAnnotate Desktop](https://opencv.org/superannotate-desktop/) 325 | - [Curve-GCN](https://github.com/fidler-lab/curve-gcn) 326 | - [VoTT (Visual Object Tagging Tool)](https://github.com/microsoft/VoTT) 327 | - [MakeSense AI](https://www.makesense.ai/) 328 | - [Awesome Data Labeling](https://github.com/heartexlabs/awesome-data-labeling) 329 | 330 | ## Building Datasets: 331 | - [Building Image Datasets for Computer Vision Algorithms](https://blog.paperspace.com/building-computer-vision-datasets/) 332 | - [icrawler - mini image framework of web crawlers](https://github.com/hellock/icrawler) [_Great_] 333 | - [Unity Computer Vision](https://github.com/Unity-Technologies/com.unity.perception) 334 | - [pytube - downloading YouTube Videos](https://github.com/pytube/pytube) 335 | - [Subreddit Media Downloader](https://github.com/Psycoguana/SubredditMediaDownloader) 336 | - [pigeon: Quickly annotate data on Jupyter](https://github.com/agermanidis/pigeon) 337 | 338 | ## LLMs & Generative AI in Vision Tasks: 339 | - [Draw UI via Generative AI](https://www.linkedin.com/posts/eric-vyacheslav-156273169_this-will-change-the-whole-industry-draw-a-ui-activity-7130527961169969153-JAi5?utm_source=share&utm_medium=member_desktop) 340 | - [Emu video & Emu edit meta's Models](https://www.linkedin.com/posts/aiatmeta_today-were-announcing-two-new-milestones-activity-7130948419409375232-7sAh?utm_source=share&utm_medium=member_desktop) 341 | 342 | 343 | ## NSFW (Not Safe for Work): 344 | - [Fine-Tuned Vision Transformer (ViT) for NSFW Image Classification](https://huggingface.co/Falconsai/nsfw_image_detection) 345 | 346 | ## Soccer & Football AI code: 347 | - [Football AI code](https://www.linkedin.com/posts/skalskip92_computervision-objectdetection-tracking-activity-7222143008358453248-1MY8?utm_source=share&utm_medium=member_desktop) 348 | - [Football AI](https://x.com/skalskip92/status/1826693515189125433) 349 | 350 | ## Other: 351 | 352 | ### Color Channel Order: 353 | - OpenCV is **BGR**, Pillow is **RGB**, and Decord is **RGB** 354 | - [Numpy Image bgr-to-rgb](https://www.scivision.dev/numpy-image-bgr-to-rgb/) 355 | 356 | - [Imagenet 1000 class indices to human readable labels](https://gist.github.com/yrevar/942d3a0ac09ec9e5eb3a) 357 | 358 | - [streamlit Make Images & Faces as Comic one!!](https://share.streamlit.io/nathannguyen-dev/comic_me_v1/main.py) 359 | - [How DALL-E 2 could solve major computer vision challenges](https://venturebeat.com/2022/04/16/how-dall-e-2-could-solve-major-computer-vision-challenges/) 360 | - [PyTorch jpeg Decoding on the GPU](https://pytorch.org/blog/pytorch-1.9-new-library-releases/#beta-jpeg-decoding-on-the-gpu) 361 | 362 | ### Miscellaneous: 363 | - [An image speaks a thousand words, but can everyone listen? On image transcreation for cultural relevance](https://arxiv.org/abs/2404.01247) 364 | -------------------------------------------------------------------------------- /DataScience.md: -------------------------------------------------------------------------------- 1 | # Great Deep Learning Tutorials for Data Science 2 | A Great Collection of Deep Learning Tutorials and Repositories for Data Science 3 | 4 | ## General: 5 | - [Python Data Science Tutorials](https://realpython.com/tutorials/data-science/) 6 | - [Data Science Course](https://github.com/amingheibi/Data-Science-Course) 7 | - [TensorFlow Decision Forests](https://github.com/tensorflow/decision-forests) 8 | - [First Steps With PySpark and Big Data Processing](https://realpython.com/pyspark-intro/) 9 | - [A Brief Introduction to PySpark](https://towardsdatascience.com/a-brief-introduction-to-pyspark-ff4284701873) 10 | - [Introduction to Anomaly Detection in Python](https://blog.floydhub.com/introduction-to-anomaly-detection-in-python/) 11 | - [Time Series basics Exploring](https://www.kaggle.com/jagangupta/time-series-basics-exploring-traditional-ts) 12 | - [Understanding Variational Autoencoders (VAEs)](https://towardsdatascience.com/understanding-variational-autoencoders-vaes-f70510919f73) 13 | - [Variational Autoencoders](https://www.jeremyjordan.me/variational-autoencoders/) 14 | - [Benchmarking Performance and Scaling of Python Clustering Algorithms](https://hdbscan.readthedocs.io/en/latest/performance_and_scalability.html) [_Important_] 15 | - [Comparing Python Clustering Algorithms](https://hdbscan.readthedocs.io/en/latest/comparing_clustering_algorithms.html) 16 | - [Modular Active Learning framework](https://github.com/modAL-python/modAL) 17 | - [pycaret: An open-source, low-code machine learning library in Python](https://github.com/pycaret/pycaret) [Good] 18 | - [Lazy Predict](https://github.com/shankarpandala/lazypredict) [Good] 19 | - [AutoXGB: XGBoost + Optuna](https://github.com/abhishekkrthakur/autoxgb) [Good] 20 | - [shap-hypetune: Hyperparameters Tuning and Features Selection for Gradient Boosting Models](https://github.com/cerlymarco/shap-hypetune) 21 | - [Deepnote: Great data science notebook](https://deepnote.com/) 22 | - [khanacademy statistics course](https://www.khanacademy.org/math/ap-statistics) [_Good_] 23 | - [NumPy Exercises](https://www.w3resource.com/python-exercises/numpy/index.php) [_Good_] 24 | - [Data Engineering Zoomcamp](https://github.com/DataTalksClub/data-engineering-zoomcamp) 25 | - [Vaex: high performance Python library for DataFrames](https://github.com/vaexio/vaex) [Great] 26 | 27 | ## The best exploratory data analysis (EDA) Tools: 28 | - [pandas-profiling](https://github.com/pandas-profiling/pandas-profiling) 29 | - [sweetviz](https://github.com/fbdesignpro/sweetviz) 30 | - [DataPrep: The easiest way to prepare data in Python](https://dataprep.ai/) 31 | - [Speeding Up Exploratory Data Analysis with Python](https://towardsdatascience.com/speeding-up-exploratory-data-analysis-with-python-838fe5e25b43) 32 | 33 | ## Scikit-learn Algorithms on GPU & for Large-Scale Data Sets: 34 | - [skorch - scikit-learn compatible neural network library that wraps PyTorch](https://github.com/skorch-dev/skorch) 35 | - [scikit-cuda](https://github.com/lebedov/scikit-cuda) 36 | - [Hummingbird - trained traditional ML models into tensor computations](https://github.com/microsoft/hummingbird) 37 | 38 | ## Anomaly Detection: 39 | - [An Awesome Tutorial to Learn Outlier Detection in Python using PyOD Library](https://www.analyticsvidhya.com/blog/2019/02/outlier-detection-python-pyod/) 40 | - [PyOD: Python Outlier Detection](https://github.com/yzhao062/pyod) [**Great**] 41 | - [PyNomaly](https://github.com/vc1492a/PyNomaly) 42 | - [DeepLog: Anomaly detection and diagnosis from system logs through deep learning](https://github.com/Thijsvanede/DeepLog) 43 | 44 | ## Bayesian Neural Networks (BNNs) & Bayesian Recommendar System: 45 | - [TyXe: Pyro-based BNNs for Pytorch users](https://github.com/TyXe-BDL/TyXe) 46 | - [Bayesian Personalized Ranking](https://github.com/shah314/BPR) 47 | - [PYROS: PYthon RecOmmender Systems library](https://github.com/makgyver/pyros) 48 | - [One-class Bayesian Personalized Ranking](https://github.com/YifanZhou95/Bayesian-Personalized-Ranking) 49 | 50 | ## Modeling Numerical & Categorical Data in ML: 51 | - [How to Prepare Categorical Data for Deep Learning in Python](https://machinelearningmastery.com/how-to-prepare-categorical-data-for-deep-learning-in-python/) 52 | - [Handling Categorical Data in ML Models](https://www.pluralsight.com/guides/handling-categorical-data-in-machine-learning-models) 53 | - [Encoding Categorical Data](https://www.analyticsvidhya.com/blog/2020/08/types-of-categorical-data-encoding/) 54 | ## Feature Engineering: 55 | - [Lime: Explaining the predictions of any machine learning classifier](https://github.com/marcotcr/lime) 56 | - [Lime Tutorial: Building Trust in Machine Learning Models (using LIME in Python)](https://www.analyticsvidhya.com/blog/2017/06/building-trust-in-machine-learning-models/) 57 | - [Missing Values: End-to-End Introduction to Handling Missing Values](https://www.analyticsvidhya.com/blog/2021/10/end-to-end-introduction-to-handling-missing-values/) 58 | 59 | ## Click-Through Rate (CTR) Prediction: 60 | - [DeepCTR](https://github.com/shenweichen/DeepCTR) 61 | 62 | ## Dynamic Time Warping (Time Series): 63 | - [Dynamic Time Warping](https://en.wikipedia.org/wiki/Dynamic_time_warping) 64 | 65 | ## Data Science Courses: 66 | - [Data Science Course](https://github.com/briandalessandro/DataScienceCourse) 67 | 68 | ## Working with Tabular Data: 69 | - [Yandex RTDL Library](https://github.com/yandex-research/rtdl) 70 | - [Vaex: high performance Python library for DataFrames](https://github.com/vaexio/vaex) [Great] 71 | 72 | ## Data Science Interview Notes: 73 | - [Cracking the Data Science Interview](https://github.com/khanhnamle1994/cracking-the-data-science-interview) 74 | - [Data Engineer Interview Questions Python](https://realpython.com/data-engineer-interview-questions-python/) 75 | - [Facebook Data Scientist Interview Questions](https://www.glassdoor.com/Interview/Facebook-Data-Scientist-Interview-Questions-EI_IE40772.0,8_KO9,23.htm) 76 | - [Interview Guides: Facebook Data Scientist](https://prepfully.com/interview-guides/facebook-data-scientist) 77 | - [Meta (Facebook) Data Science Interview Questions and Solutions](https://www.interviewquery.com/blog-facebook-data-science-interview-questions-and-solutions) 78 | 79 | ## Deep Learning Recommendation Model: 80 | - [Introducing TorchRec](https://pytorch.org/blog/introducing-torchrec/) 81 | - [TorchRec](https://github.com/pytorch/torchrec) 82 | - [Deep Learning Recommendation Model for Personalization and Recommendation Systems - DLRM](https://github.com/facebookresearch/dlrm) 83 | - [DLRM: An advanced, open source deep learning recommendation model](https://ai.facebook.com/blog/dlrm-an-advanced-open-source-deep-learning-recommendation-model/) 84 | - [LightFM](https://github.com/lyst/lightfm) 85 | - [Neural Recommendation Algorithms](https://towardsdatascience.com/recotour-ii-neural-recommendation-algorithms-49733938d56e) 86 | - [Build a Recommendation Engine With Collaborative Filtering](https://realpython.com/build-recommendation-engine-collaborative-filtering/) [_**Great**_] 87 | - [NCF - Neural Collaborative Filtering](https://github.com/NervanaSystems/distiller/tree/master/examples/ncf) 88 | - [Using Neural Networks for your Recommender System](https://developer.nvidia.com/blog/using-neural-networks-for-your-recommender-system/) [**Great**] 89 | - [Neural Collaborative Filtering](https://towardsdatascience.com/neural-collaborative-filtering-96cef1009401) 90 | - [AWS Personalized Recommendation Model](https://aws.amazon.com/personalize/) 91 | - [Microsoft Recommenders](https://github.com/microsoft/recommenders) [Great] 92 | - [Monolith: Real Time Recommendation System of TikTok](https://arxiv.org/pdf/2209.07663.pdf) 93 | - [Twitter's Recommendation Algorithm](https://github.com/twitter/the-algorithm-ml) 94 | - [TikTok recommender system Notes](https://www.linkedin.com/posts/damienbenveniste_machinelearning-datascience-artificialintelligence-activity-7050129099549249536-TctQ/?utm_source=share&utm_medium=member_android) 95 | - [Monolith: Real Time Recommendation System With Collisionless Embedding Table](https://arxiv.org/abs/2209.07663) 96 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2019 Amir Hossein Karami 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /NLP.md: -------------------------------------------------------------------------------- 1 | # Great Deep Learning Tutorials for Natural Language Processing (NLP) 2 | A Great Collection of Deep Learning Tutorials and Repositories for Natural Language Processing (NLP) 3 | 4 | ## General: 5 | - [Great NLP Posts](http://jalammar.github.io/) 6 | - [Awesome NLP Paper Discussions - Hugging Face](https://github.com/huggingface/awesome-papers) [_Excellent_] 7 | - [Ten trends in Deep learning NLP](https://blog.floydhub.com/ten-trends-in-deep-learning-nlp/) 8 | - [Attention in RNNs](https://medium.com/datadriveninvestor/attention-in-rnns-321fbcd64f05) 9 | - [Understanding self-attention and other types of attention mechanisms](https://www.linkedin.com/posts/sebastianraschka_understanding-and-coding-self-attention-activity-7152300807080546304-uu21?utm_source=share&utm_medium=member_desktop) 10 | - [BERT - TensorFlow](https://github.com/google-research/bert) 11 | - [Understanding XLNet](https://www.borealisai.com/en/blog/understanding-xlnet/) 12 | - [XLNet - TensorFlow](https://github.com/zihangdai/xlnet) 13 | - [XLM (PyTorch implementation of Cross-lingual Language Model Pretraining)](https://github.com/facebookresearch/XLM) 14 | - [Pretrained PyTorch models for BERT](https://github.com/huggingface/pytorch-pretrained-BERT) 15 | - [Library of state-of-the-art pretrained models for NLP](https://github.com/huggingface/pytorch-transformers#quick-tour) [_Excellent_] 16 | - [DistilBERT](https://medium.com/huggingface/distilbert-8cf3380435b5) 17 | - [FastBert](https://arxiv.org/abs/2311.10770) 18 | - [FastBert Linkedin Post](https://www.linkedin.com/posts/activity-7132888497119485952-GMsV?utm_source=share&utm_medium=member_desktop) 19 | - [PyTorch Hub - BERT](https://pytorch.org/hub/huggingface_pytorch-pretrained-bert_bert/) 20 | - [A Simple Guide On Using BERT for Binary Text Classification](https://medium.com/swlh/a-simple-guide-on-using-bert-for-text-classification-bbf041ac8d04) 21 | - [Core ML 3 implementation of BERT for Question answering](https://github.com/huggingface/swift-coreml-transformers) 22 | - [NLP - Keras - Intro](https://nlpforhackers.io/keras-intro/) 23 | - [AllenNLP](https://allennlp.org/) [_General NLP_] 24 | - [Stanza - A Python NLP Library for Many Human Languages](https://stanfordnlp.github.io/stanza/) 25 | - [The Best NLP Papers From ICLR 2020](https://www.topbots.com/best-nlp-papers-from-iclr-2020) 26 | - [Deep learning for natural language processing and information retrieval at the University of Waterloo](https://github.com/castorini) 27 | - [Natural Language Processing With spaCy in Python](https://realpython.com/natural-language-processing-spacy-python/) [_Great_] 28 | - [NLP Papers](https://github.com/AliAkbarBadri/nlp-papers) 29 | - [A Great NLP Course](https://lena-voita.github.io/nlp_course.html) 30 | - [KerasNLP: Modular NLP Workflows for Keras](https://github.com/keras-team/keras-nlp) 31 | - [NLP Test: Deliver Safe & Effective Models](https://github.com/JohnSnowLabs/nlptest) 32 | - [Karpathy minbpe](https://github.com/karpathy/minbpe) 33 | - [Karpathy's 2 Hours Tutorial for Building GPT Tokenizer](https://www.linkedin.com/posts/liorsinclair_andrej-karpathy-just-uploaded-a-new-2-hour-activity-7165765602492571650-io92?utm_source=share&utm_medium=member_desktop) 34 | - [Learning Core Foundational Concepts in NLP by Examples and by calculation by Hand](https://www.linkedin.com/posts/alphasignal_can-foundational-concepts-like-transformers-activity-7163890641054232576-B1ai?utm_source=share&utm_medium=member_android) 35 | - [SetFit: Efficient Few-shot Learning with Sentence Transformers](https://github.com/huggingface/setfit) 36 | 37 | ## General Persian based libraries & Data Sets: 38 | - [Parsivar: library for Persian text preprocessing](https://github.com/ICTRC/Parsivar) 39 | - [Hazm](https://github.com/sobhe/hazm) 40 | - [persianNLP](https://github.com/persiannlp) 41 | - [ParsiNLU: Comprehensive suit of high-level NLP tasks for Persian language](https://github.com/persiannlp/parsinlu) 42 | - [FarsTail: A Persian Natural Language Inference Dataset](https://github.com/dml-qom/FarsTail) 43 | - [wordfreq: Access a database of word frequencies](https://github.com/rspeer/wordfreq) 44 | - [Persian Stop Words List](https://github.com/kharazi/persian-stopwords) 45 | - [Persian Stop Words List in Hazm Repo](https://github.com/sobhe/hazm/blob/master/hazm/data/stopwords.dat) 46 | - [PCoQA: Persian Conversational Question Answering Dataset](https://github.com/HamedHematian/PCoQA) 47 | - [Khayyam Challenge (PersianMMLU): Is Your LLM Truly Wise to The Persian Language?](https://arxiv.org/html/2404.06644v1) [Good paper & dataset] 48 | - [Basalam Dataset via RadeAI Team](https://www.linkedin.com/posts/rade-ai_datascience-machinelearning-basalam-activity-7193561781280157696-NF8T?utm_source=share&utm_medium=member_desktop) 49 | - [Basalam Datasets for LLM Fine-tuning](https://www.linkedin.com/posts/mohammadreza-esmaeilian-572ba9193_%D8%A7%D9%86%D8%AA%D8%B4%D8%A7%D8%B1-%D8%AF%DB%8C%D8%AA%D8%A7%D8%B3%D8%AA%D9%87%D8%A7-%D9%88-llm%D9%87%D8%A7%DB%8C-%D9%81%D8%A7%DB%8C%D9%86%D8%AA%DB%8C%D9%88%D9%86-%D8%B4%D8%AF%D9%87-%D8%A7%D8%AE%D8%AA%D8%B5%D8%A7%D8%B5%DB%8C-activity-7204220860142989314-VDUO?utm_source=share&utm_medium=member_desktop) 50 | - [ParsBench](https://www.linkedin.com/posts/shahriarshm_llm-dataset-syntheticabrdataset-activity-7278063501909098496-KR0O?utm_source=share&utm_medium=member_desktop) 51 | 52 | ## Text Representation: 53 | - [Beyond Word Embeddings Part 1](https://towardsdatascience.com/beyond-word-embeddings-part-1-an-overview-of-neural-nlp-milestones-82b97a47977f) 54 | - [Beyond Word Embeddings Part 2](https://towardsdatascience.com/beyond-word-embeddings-part-2-word-vectors-nlp-modeling-from-bow-to-bert-4ebd4711d0ec) 55 | - [Learning Word Embedding](https://lilianweng.github.io/lil-log/2017/10/15/learning-word-embedding.html) 56 | - [Introduction to Word Embedding and Word2Vec](https://towardsdatascience.com/introduction-to-word-embedding-and-word2vec-652d0c2060fa) 57 | - [Word Embedding](https://medium.com/data-science-group-iitr/word-embedding-2d05d270b285) 58 | - [Understanding Word Embeddings](https://hackernoon.com/understanding-word-embeddings-a9ff830403ce) 59 | - [Introduction to Word Vectors](https://medium.com/@jayeshbahire/introduction-to-word-vectors-ea1d4e4b84bf) 60 | - [Word2vec Made Easy](https://towardsdatascience.com/word2vec-made-easy-139a31a4b8ae) 61 | - [What is GloVe? Part I](https://towardsdatascience.com/emnlp-what-is-glove-part-i-3b6ce6a7f970) 62 | - [What is GloVe? Part II](https://towardsdatascience.com/emnlp-what-is-glove-part-ii-9e5ad227ee0) 63 | - [What is GloVe? Part III](https://towardsdatascience.com/emnlp-what-is-glove-part-iii-c6090bed114) 64 | - [What is GloVe? Part IV](https://towardsdatascience.com/emnlp-what-is-glove-part-iv-e605a4c407c8) 65 | - [What is GloVe? Part V](https://towardsdatascience.com/emnlp-what-is-glove-part-v-fa888272c290) 66 | - [ELMo: Deep Contextualized Word Representation](https://allennlp.org/elmo) 67 | - [A Step-by-Step NLP Guide to Learn ELMo](https://www.analyticsvidhya.com/blog/2019/03/learn-to-use-elmo-to-extract-features-from-text/) 68 | - [ELMo: Contextual language embedding](https://towardsdatascience.com/elmo-contextual-language-embedding-335de2268604) 69 | - [word embeddings with ELMo](https://medium.com/saarthi-ai/elmo-for-contextual-word-embedding-for-text-classification-24c9693b0045) 70 | - [Doc2Vec - Gensim](https://radimrehurek.com/gensim/models/doc2vec.html) 71 | 72 | ## Self-Supervised Learning in NLP: 73 | - [https://amitness.com/2020/05/self-supervised-learning-nlp/](https://amitness.com/2020/05/self-supervised-learning-nlp/) 74 | - [COSINE: Fine-Tuning Pre-trained Language Model with Weak Supervision](https://github.com/yueyu1030/COSINE) 75 | 76 | ## RNN, LSTM, and GRU: 77 | - [Understanding LSTM Networks](https://colah.github.io/posts/2015-08-Understanding-LSTMs/) 78 | - [Illustrated Guide to LSTM’s and GRU’s](https://towardsdatascience.com/illustrated-guide-to-lstms-and-gru-s-a-step-by-step-explanation-44e9eb85bf21) 79 | - [Animated RNN, LSTM and GRU](https://towardsdatascience.com/animated-rnn-lstm-and-gru-ef124d06cf45) 80 | - [Recurrent Neural Networks and LSTM explained](https://medium.com/@purnasaigudikandula/recurrent-neural-networks-and-lstm-explained-7f51c7f6bbb9) 81 | - [Long Short-Term Memory (LSTM): Concept](https://medium.com/@kangeugine/long-short-term-memory-lstm-concept-cb3283934359) 82 | - [Understanding architecture of LSTM cell from scratch](https://hackernoon.com/understanding-architecture-of-lstm-cell-from-scratch-with-code-8da40f0b71f4) 83 | - [Basic understanding of LSTM](https://blog.goodaudience.com/basic-understanding-of-lstm-539f3b013f1e) 84 | - [Taming LSTMs with PyTorch](https://towardsdatascience.com/taming-lstms-variable-sized-mini-batches-and-why-pytorch-is-good-for-your-health-61d35642972e) 85 | - [Introduction to LSTM](https://www.analyticsvidhya.com/blog/2017/12/fundamentals-of-deep-learning-introduction-to-lstm/?utm_medium=ELMoNLParticle&utm_source=blog) 86 | - [Introduction to RNNs](https://www.jeremyjordan.me/introduction-to-recurrent-neural-networks/) 87 | - [xLSTM - Post1](https://www.linkedin.com/posts/liorsinclair_is-this-the-end-of-transformers-the-team-activity-7194350205318701056-8yBr?utm_source=share&utm_medium=member_desktop) 88 | - [Were RNNs All We Needed?](https://arxiv.org/abs/2410.01201) [Interesting Paper] 89 | 90 | ## Transformers: 91 | - [How Transformers Work](https://towardsdatascience.com/transformers-141e32e69591) 92 | - [The Illustrated Transformer](http://jalammar.github.io/illustrated-transformer/) 93 | - [Transformers from Scratch](https://e2eml.school/transformers.html) 94 | - [What is a Transformer?](https://medium.com/inside-machine-learning/what-is-a-transformer-d07dd1fbec04) 95 | - [How Transformers work in deep learning and NLP](https://theaisummer.com/transformer/) 96 | - [Transformer: A Novel Neural Network Architecture for Language Understanding](https://ai.googleblog.com/2017/08/transformer-novel-neural-network.html) 97 | - [How do Transformers Work in NLP?](https://www.analyticsvidhya.com/blog/2019/06/understanding-transformers-nlp-state-of-the-art-models/) 98 | - [The Essence of Transformers](https://towardsdatascience.com/the-essence-of-transformers-9fb8e14cc465) [Good] 99 | - [Transformers and Multi Head Attention](https://uvadlc-notebooks.readthedocs.io/en/latest/tutorial_notebooks/tutorial6/Transformers_and_MHAttention.html) 100 | - [Multi Head Attention](https://d2l.ai/chapter_attention-mechanisms-and-transformers/multihead-attention.html) 101 | - [BERT for Dummies](https://towardsdatascience.com/bert-for-dummies-step-by-step-tutorial-fb90890ffe03) 102 | - [The Dark Secrets of BERT](https://text-machine-lab.github.io/blog/2020/bert-secrets/) 103 | - [A Survey of Long-Term Context in Transformers](https://www.pragmatic.ml/a-survey-of-methods-for-incorporating-long-term-context/) [_Great_] 104 | - [The Transformer Family](https://lilianweng.github.io/lil-log/2020/04/07/the-transformer-family.html) 105 | - [The Transformer Isn’t As Hard To Understand As You Might Think](https://towardsdatascience.com/knocking-on-transformers-door-attention-mechanism-explained-intuitively-df5d4fcecdf8) 106 | - [Review of Compact Transformer Architectures](https://medium.com/@jfd2139/review-of-compact-transformer-architectures-c477b797e2d5) [**Great**] 107 | - [REFORMER: The Efficient Transformer](https://arxiv.org/pdf/2001.04451.pdf) 108 | - [GPT-3: Language Models are Few-Shot Learners](https://github.com/openai/gpt-3) 109 | - [GPT-3 Sandbox](https://github.com/shreyashankar/gpt3-sandbox) 110 | - [Microsoft will launch GPT-4](https://medium.com/@yablonassaf/microsoft-will-launch-gpt-4-with-ai-videos-on-wednesday-75d882e0260e) 111 | - [OpenAI GPT-4](https://openai.com/research/gpt-4) 112 | - [Some information about GPT-4](https://www.linkedin.com/posts/damienbenveniste_machinelearning-datascience-artificialintelligence-activity-7041793426530390016-5P-n/?utm_source=share&utm_medium=member_android) 113 | - [Regular Expressions (Regex) Generated by GPT-3](https://losslesshq.com/) 114 | - [Auto Regex: Converting English description to Regex](https://www.autoregex.xyz/) [Good] 115 | - [minGPT](https://github.com/karpathy/minGPT) 116 | - [NVIDIA FasterTransformer: Transformer related optimization, including BERT & GPT](https://github.com/NVIDIA/FasterTransformer) 117 | - [OpenNMT CTranslate2: Fast inference engine for Transformer models](https://github.com/OpenNMT/CTranslate2/) 118 | - [Deploying GPT-J and T5 with FasterTransformer and Triton Inference Server](https://developer.nvidia.com/blog/deploying-gpt-j-and-t5-with-fastertransformer-and-triton-inference-server/?ncid=so-link-499508#cid=dl05_so-link_en-us) [Interesting] 119 | - [MEND: Fast Model Editing at Scale](https://github.com/eric-mitchell/mend) [**Excellent Work**] 120 | - [BorealisAI Transformers I: Introduction](https://www.borealisai.com/research-blogs/tutorial-14-transformers-i-introduction/) 121 | - [OpenAI Best Practices for Deploying Language Models](https://openai.com/blog/best-practices-for-deploying-language-models/) 122 | - [OPT-IML](https://github.com/facebookresearch/metaseq/tree/main/projects/OPT-IML) 123 | - [RetNet: an Alternative to Transformers](https://www.linkedin.com/posts/aleksagordic_an-alternative-to-transformers-whoa-activity-7087790555190980608-66ZM?utm_source=share&utm_medium=member_android) 124 | - [Transformer Taxonomy](https://kipp.ly/blog/transformer-taxonomy/) [Great] 125 | - [Generative AI exists because of the transformer: Great Visual Explanation](https://ig.ft.com/generative-ai/) [Great] 126 | 127 | ### Reinforcement Learning from Human Feedback (RLHF): 128 | - [RLHF Tutorial](https://vinija.ai/concepts/RLHF/) 129 | - [New method instead of RLHF: Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://www.linkedin.com/posts/yoelzeldes_to-get-llms-as-good-as-openais-gpt-4-is-activity-7078958558519656451-N6Wo/?utm_source=share&utm_medium=member_android) 130 | - [Finetuning an LLM: RLHF and alternatives (Part I)](https://argilla.io/blog/mantisnlp-rlhf-part-1/) 131 | - [Finetuning an LLM: RLHF and alternatives (Part II)](https://argilla.io/blog/mantisnlp-rlhf-part-2/) 132 | - [Finetuning an LLM: RLHF and alternatives (Part III)](https://argilla.io/blog/mantisnlp-rlhf-part-3/) 133 | - [How good is AI feedback?](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_how-good-is-ai-feedback-and-does-it-really-activity-7171174171413102592-eVs9?utm_source=share&utm_medium=member_desktop) 134 | - [Direct Preference Optimization (DPO) for LLM Alignment (From Scratch)](https://github.com/rasbt/LLMs-from-scratch/blob/main/ch07/04_preference-tuning-with-dpo/dpo-from-scratch.ipynb) 135 | 136 | ### Tokenizer Notes: 137 | - [𝗻𝗲𝘄 𝗽𝗮𝗽𝗲𝗿 𝗯𝘆 𝗠𝗲𝘁𝗮 𝗰𝗹𝗮𝗶𝗺𝘀 𝘁𝗵𝗮𝘁 𝘄𝗲 𝗰𝗮𝗻 𝗴𝗲𝘁 𝗿𝗶𝗱 𝗼𝗳 𝘁𝗼𝗸𝗲𝗻𝗶𝘇𝗲𝗿𝘀: Byte Latent Transformer: Patches Scale Better Than Tokens --> we could get rid of tokenizers](https://www.linkedin.com/posts/a-roucher_%F0%9D%97%A3%F0%9D%97%BC%F0%9D%98%81%F0%9D%97%B2%F0%9D%97%BB%F0%9D%98%81%F0%9D%97%B6%F0%9D%97%AE%F0%9D%97%B9-%F0%9D%97%BD%F0%9D%97%AE%F0%9D%97%BF%F0%9D%97%AE%F0%9D%97%B1%F0%9D%97%B6%F0%9D%97%B4%F0%9D%97%BA-%F0%9D%98%80%F0%9D%97%B5%F0%9D%97%B6%F0%9D%97%B3%F0%9D%98%81-activity-7273382398891810816-QfQo?utm_source=share&utm_medium=member_desktop) 138 | - [Byte Latent Transformer: Patches Scale Better Than Tokens (paper)](https://dl.fbaipublicfiles.com/blt/BLT__Patches_Scale_Better_Than_Tokens.pdf) 139 | 140 | ### Large Language Models (LLMs): 141 | - [LLM Reading Papers](https://www.linkedin.com/posts/eric-vyacheslav-156273169_new-must-read-the-anti-hype-llm-reading-activity-7247244292568625152-DQsb?utm_source=share&utm_medium=member_desktop) 142 | - [LLaMA](https://github.com/facebookresearch/llama) 143 | - [Toolformer: Language Models Can Teach Themselves to Use Tools](https://arxiv.org/abs/2302.04761) [Great] 144 | - [Toolformer GitHub](https://github.com/lucidrains/toolformer-pytorch) 145 | - [Amazon Multimodal Chain-of-Thought Reasoning in Language Models](https://github.com/amazon-science/mm-cot) 146 | - [LLaMA-based ChatGPT Training](https://github.com/nebuly-ai/nebullvm/tree/main/apps/accelerate/chatllama) [Great] 147 | - [The Wisdom of Hindsight Makes Language Models Better Instruction Followers](https://arxiv.org/abs/2302.05206) 148 | - [Stanford Alpaca: An Instruction-following LLaMA model](https://github.com/tatsu-lab/stanford_alpaca) 149 | - [Alpaca: A Strong, Replicable Instruction-Following Model](https://crfm.stanford.edu/2023/03/13/alpaca.html) 150 | - [Fine-Tune Alpaca in Arabic](https://www.linkedin.com/posts/yassine-boukhari-006748217_alpaca-a-strong-replicable-instruction-following-activity-7043223149710036992-YUJb?utm_source=share&utm_medium=member_android) 151 | - [TRL: Transformer Reinforcement Learning](https://github.com/lvwerra/trl) 152 | - [Large Language Model (LLM) Primers Tutorial](https://www.linkedin.com/posts/amanc_artificialintelligence-machinelearning-ai-activity-7045245910850695168-Fp9K/?utm_source=share&utm_medium=member_android) [Great] 153 | - [Dolly](https://www.databricks.com/blog/2023/03/24/hello-dolly-democratizing-magic-chatgpt-open-models.html) 154 | - [Microsoft JARVIS & HuggingGPT](https://github.com/microsoft/JARVIS) [Interesting] 155 | - [open-source LLMs](https://www.linkedin.com/posts/sahar-mor_artificialintelligence-machinelearning-activity-7049789761728770049-QLsv/?utm_source=share&utm_medium=member_android) 156 | - [GPT4Free](https://github.com/xtekky/gpt4free) 157 | - [HuggingChat](https://huggingface.co/chat/) 158 | - [LaMini-LM: A Diverse Herd of Distilled Models](https://github.com/mbzuai-nlp/LaMini-LM/) 159 | - [RedPajama-Data: An Open Source Recipe to Reproduce LLaMA training dataset](https://github.com/togethercomputer/RedPajama-Data) 160 | - [BigCode](https://huggingface.co/bigcode) 161 | - [OpenLLaMA](https://github.com/openlm-research/open_llama) 162 | - [Dromedary: towards helpful, ethical and reliable LLMs](https://github.com/IBM/Dromedary) 163 | - [MPT-7B Model with Commercial Licence](https://huggingface.co/mosaicml/mpt-7b/blob/main/README.md) 164 | - [MPT-7B Story Writer](https://huggingface.co/mosaicml/mpt-7b-storywriter) 165 | - [MPT-7B](https://github.com/mosaicml/llm-foundry) 166 | - [MPT-7B Blog](https://www.mosaicml.com/blog/mpt-7b) 167 | - [Open LLMs](https://github.com/eugeneyan/open-llms) 168 | - [Google PaLM 2](https://ai.google/discover/palm2) 169 | - [BLOOMChat](https://github.com/sambanova/bloomchat) 170 | - [LLMs Practical Guide](https://github.com/Mooler0410/LLMsPracticalGuide) 171 | - [FrugalGPT](https://www.linkedin.com/posts/sanyambhutani_saving-98-llm-usage-costs-stanford-activity-7062420577357037568-t0a8/?utm_source=share&utm_medium=member_android) 172 | - [ChatALL](https://github.com/sunner/ChatALL) [Great] 173 | - [Falcon LLM](https://falconllm.tii.ae/) 174 | - [The Falcon has landed in the Hugging Face ecosystem](https://huggingface.co/blog/falcon) [Great] 175 | - [Open LLMs](https://github.com/eugeneyan/open-llms) [Great] 176 | - [OpenLLMs: Less is More for Open-source Models](https://github.com/imoneoi/openchat) [Great] 177 | - [LLaMA2](https://www.llama2.ai/) 178 | - [source code of llama2-chatbot](https://github.com/a16z-infra/llama2-chatbot/tree/main) 179 | - [Notes about OpenAI's GPT-4 Model](https://www.linkedin.com/posts/aleksagordic_openais-gpt-4-details-have-apparently-been-activity-7085226267712614400-T1d3/?utm_source=share&utm_medium=member_android) 180 | - [GPT-4 is getting worse over time](https://www.linkedin.com/posts/svpino_gpt-4-is-getting-worse-over-time-not-better-activity-7087379892077481984-uORp/?utm_source=share&utm_medium=member_android) 181 | - [OpenChat: Less is More for Open-source Models](https://huggingface.co/openchat/openchat) 182 | - [Instruction Tuning Datasets](https://github.com/raunak-agarwal/instruction-datasets) 183 | - [ToolLLM](https://www.linkedin.com/posts/omarsar_enabling-llms-with-tool-use-capabilities-activity-7093299751571320832-1WHU/?utm_source=share&utm_medium=member_android) 184 | - [Falcon 180B](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_falcon-180b-released-tii-just-released-activity-7105166508376367105-P7ws?utm_source=share&utm_medium=member_desktop) 185 | - [Fine-tune Falcon 180B using QLoRA and Flash Attention on Amazon SageMaker](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_fine-tune-falcon-180b-with-qlora-and-flash-activity-7107387875515580416-zhSe?utm_source=share&utm_medium=member_desktop) 186 | - [Large Language Models as Optimizers](https://arxiv.org/abs/2309.03409) 187 | - [Favourite LLM Authors](https://www.linkedin.com/posts/sanyambhutani_curated-list-of-my-favourite-llm-authors-activity-7105896422226423808-Unev?utm_source=share&utm_medium=member_desktop) 188 | - [Open Source LLMs for Commercial Use](https://www.linkedin.com/posts/armand-ruiz_top-open-source-llms-available-for-commercial-activity-7137772625468002304-jkMM?utm_source=share&utm_medium=member_desktop) 189 | - [Optimizing your LLM in production](https://huggingface.co/blog/optimize-llm) [Important] 190 | - [In Context Vectors (ICV): an alternative to Few-Shot Learning and Finetuning techniques like LoRA to improve an LLMs performance](https://www.linkedin.com/posts/pramodith_in-context-vectors-icv-is-an-alternative-activity-7131970618467471360-67Z3?utm_source=share&utm_medium=member_desktop) 191 | - [NexusRavan v2 13B Fuction Calling LLM Surpassing GPT-4](https://www.linkedin.com/posts/nexusflow-ai_nexusravenv2-opensource-genai-activity-7137805301323362304-U2Pl?utm_source=share&utm_medium=member_desktop) 192 | - [Phixtral model](https://www.linkedin.com/posts/maxime-labonne_phixtral-i-made-the-first-efficient-mixture-activity-7150758415961620481-v0qx?utm_source=share&utm_medium=member_desktop) 193 | - [Eagle-7B LLM: 100% attention-free RNN Model!](https://www.linkedin.com/posts/maxime-labonne_rwkv-released-eagle-7b-its-an-llm-that-activity-7157700712330661888-cdd1?utm_source=share&utm_medium=member_desktop) 194 | - [Eagle-7B LLM: Blog Post](https://blog.rwkv.com/p/eagle-7b-soaring-past-transformers) 195 | - [Can LLMs improve themselves? Self-play fine-tuning (SPIN)](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_can-llms-improve-themselves-self-play-fine-tuning-activity-7150501901665542144-mk4K?utm_source=share&utm_medium=member_desktop) 196 | - [AI2 OLMo Model: Linkedin Post](https://www.linkedin.com/posts/natolambert_allenaiolmo-7b-hugging-face-activity-7158834284689035264-vfu7?utm_source=share&utm_medium=member_desktop) 197 | - [AI2 OLMo Model: HuggingFace](https://huggingface.co/allenai/OLMo-7B) 198 | - [AI2 OLMo Model: Original Blog post](https://www.interconnects.ai/p/olmo) 199 | - [Some Notes about OLMo Model](https://www.linkedin.com/posts/sebastianraschka_ive-been-working-with-the-1b7b-olmo-models-activity-7166067492778360832-kc3T?utm_source=share&utm_medium=member_desktop) 200 | - [Mixtral in colab](https://github.com/dvmazur/mixtral-offloading/blob/master/notebooks/demo.ipynb) [Great] 201 | - [Grok-1 LLM with 314B Size: Post1](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_elon-musk-kept-his-word-and-released-grok-activity-7175221121472983040-F7zS?utm_source=share&utm_medium=member_desktop) 202 | - [Grok-1 LLM: Post2](https://www.linkedin.com/posts/liorsinclair_big-news-grok-is-finally-open-source-with-activity-7175496738948968448--Ewx?utm_source=share&utm_medium=member_desktop) 203 | - [Grok-3 LLM from xAI](https://x.com/lmarena_ai/status/1891706264800936307) 204 | - [Grok-3 LLM from xAI - karpathy](https://x.com/karpathy/status/1891720635363254772) 205 | - [DBRX LLM](https://www.databricks.com/blog/introducing-dbrx-new-state-art-open-llm) 206 | - [DBRX LLM: Post1](https://www.linkedin.com/posts/mateizaharia_at-databricks-weve-built-an-awesome-model-activity-7178738621099769857-v4X8?utm_source=share&utm_medium=member_desktop) 207 | - [DBRX LLM: Post2](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_new-state-of-the-art-open-llm-databricks-activity-7178748050117451776-Otgg?utm_source=share&utm_medium=member_desktop) 208 | - [LLMs via Multi-Token Prediction](https://www.linkedin.com/posts/aiatmeta_new-research-from-fair-better-faster-large-activity-7194022959609438208-TH1u?utm_source=share&utm_medium=member_android) 209 | - [Test Time Computing for Open LLMs](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_how-we-implemented-test-time-computing-for-activity-7274685354895458304-elNI?utm_source=share&utm_medium=member_desktop) 210 | 211 | ### Merge LLMs: 212 | - [Linkedin Post](https://www.linkedin.com/posts/maxime-labonne_merge-large-language-models-with-mergekit-activity-7150044812337901569-3zIu?utm_source=share&utm_medium=member_android) 213 | - [Colab Notebook](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb) 214 | - [Main Github of Mergekit](https://github.com/cg123/mergekit) 215 | - [huggingface merge-models blog post](https://huggingface.co/blog/mlabonne/merge-models) 216 | - [Making the NeuralBeagle14-7B LLM Model (via Merging models and other methods)](https://www.linkedin.com/posts/maxime-labonne_heres-how-i-made-the-new-best-performing-activity-7153302680780640256-1Sv7?utm_source=share&utm_medium=member_desktop) 217 | - [Merge Large Language Models with mergekit](https://towardsdatascience.com/merge-large-language-models-with-mergekit-2118fb392b54) 218 | - [Fine-tune a Mistral-7b model with Direct Preference Optimization](https://towardsdatascience.com/fine-tune-a-mistral-7b-model-with-direct-preference-optimization-708042745aac) 219 | - [AutoMerger](https://www.linkedin.com/posts/maxime-labonne_automerger-how-i-automated-the-model-merging-activity-7172890188430454786-Djs7?utm_source=share&utm_medium=member_desktop) 220 | - [Evolutionary LLM Merging - Post1](https://www.linkedin.com/posts/maxime-labonne_evolutionary-model-merge-sakana-ai-released-activity-7176527260097597440-52JT?utm_source=share&utm_medium=member_desktop) 221 | - [Evolutionary LLM Merging - Post2](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_the-evolution-of-llms-model-merging-is-activity-7176561819933671424-NNNX?utm_source=share&utm_medium=member_desktop) 222 | - [Mixture of Experts (MoEs) Explained](https://huggingface.co/blog/moe) [Great] 223 | - [Mixture of Experts (MoEs) Papers List](https://huggingface.co/collections/osanseviero/moes-papers-reading-list-65a83f8a9aec16459920ffe0) 224 | - [Mixture of Experts (MoEs) Linkedin Post](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_mixture-of-experts-explained-activity-7179478562398187520-dbzM?utm_source=share&utm_medium=member_desktop) 225 | - [Mixture-of-Depths - Post1](https://www.linkedin.com/posts/zaiinulabideen_crazy-ai-week-mixture-of-depths-qwen15-activity-7182746449921658880-aLVO?utm_source=share&utm_medium=member_desktop) 226 | - [Mixture-of-Depths (MoD) - Post2](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_can-we-train-llms-to-allocate-flops-compute-activity-7182303286429917184-jkOm?utm_source=share&utm_medium=member_desktop) 227 | - [AutoLoRA-Merging Linkedin Post](https://www.linkedin.com/posts/zaiinulabideen_autolora-merging-ties-dare-magnitudeprune-activity-7166081059166662658-OzxA?utm_source=share&utm_medium=member_desktop) 228 | 229 | ### LLaMA2 Related Links: 230 | - [A colab gradio web UI for running Large Language Models](https://github.com/camenduru/text-generation-webui-colab) [Great] 231 | - [llama-2-7b-chat-GPTQ-4bit](https://colab.research.google.com/github/camenduru/text-generation-webui-colab/blob/main/llama-2-7b-chat-GPTQ-4bit.ipynb) 232 | - [camenduru](https://github.com/camenduru) 233 | - [llama-2 philschmid](https://www.philschmid.de/llama-2) 234 | - [fine-tuning LLMs with TRL](https://www.linkedin.com/posts/lvwerra_it-crazy-how-far-the-ml-field-has-come-when-activity-7087699813009383425-Sr1y/?utm_source=share&utm_medium=member_android) 235 | - [lora tuning peft finetuning llama2](https://huggingface.co/docs/trl/main/en/lora_tuning_peft#finetuning-llama2-model) 236 | - [LLaMA2 with PEFT](https://www.linkedin.com/posts/gante_unleash-the-true-llama-2-potential-from-day-activity-7087363261666328577-38jV/?utm_source=share&utm_medium=member_android) 237 | - [Baby LLaMA2 in C](https://github.com/karpathy/llama2.c) 238 | - [Releasing LLongMA-2 16k](https://www.linkedin.com/posts/enrico-shippole-495521b8_conceptofmindllongma-2-13b-16k-hugging-activity-7090718505183928320-DYtD/?utm_source=share&utm_medium=member_android) 239 | - [LLaMA2 API in Hugging Face Inference](https://www.linkedin.com/feed/update/urn:li:activity:7089986843839979521/?utm_source=share&utm_medium=member_android) 240 | - [LLaMA2 API in Monster API](https://monsterapi.ai/llama-2-7b-chat-api) 241 | - [LLaMA2-Accessory](https://github.com/Alpha-VLLM/LLaMA2-Accessory) 242 | - [Hermes-LLongMA-2 8k](https://www.linkedin.com/posts/enrico-shippole-495521b8_conceptofmindhermes-llongma-2-13b-8k-hugging-activity-7092178977217282049-JZB8/?utm_source=share&utm_medium=member_android) 243 | - [Training Llama 2](https://www.linkedin.com/posts/bhavsarpratik_llama2-finetuning-genai-activity-7092496767870509056-RojZ/?utm_source=share&utm_medium=member_android) 244 | - [Llama-2-7B-32K-Instruct — and fine-tuning for Llama-2 models with Together API](https://together.ai/blog/llama-2-7b-32k-instruct) 245 | - [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory) 246 | - [LLaMA-Factory Notes](https://www.linkedin.com/posts/rorcde_llama-factory-ai-library-of-the-day-llama-activity-7138958059506143234-t5p2?utm_source=share&utm_medium=member_desktop) 247 | - [Purple llama by Meta - Link1](https://github.com/facebookresearch/PurpleLlama) 248 | - [Purple llama by Meta - Link2](https://www.linkedin.com/posts/aiatmeta_announcing-purple-llama-towards-open-trust-activity-7138536031858937857-edXE?utm_source=share&utm_medium=member_desktop) 249 | - [Purple llama by Meta - Link3](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_purple-llama-just-got-released-by-meta-activity-7138538944115200001-WKAR?utm_source=share&utm_medium=member_desktop) 250 | - [TinyLLaMa-1.1B](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) 251 | - [Can llama learn new language?](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_is-it-possible-to-teach-llms-a-different-activity-7148653756165812226--l7o?utm_source=share&utm_medium=member_desktop) 252 | - [Persian LLaMa](https://huggingface.co/spaces/mostafaamiri/persianllama) 253 | 254 | ### LLaMA3 Related Links: 255 | - [LLaMA3 Linkedin Post1](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_welcome-llama-3-metas-new-open-llm-activity-7186762894989012992-SBLe?utm_source=share&utm_medium=member_desktop) 256 | - [Meta LLaMA3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) 257 | - [Fine tune LLaMA3](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_efficiently-fine-tune-llama-3-with-pytorch-activity-7188186109363859456-sYSR?utm_source=share&utm_medium=member_desktop) 258 | - [LLaMA3 Long Context](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_llama-3-extended-to-almost-100000-token-activity-7189518531300904963-9Y9V?utm_source=share&utm_medium=member_desktop) 259 | - [LLaMA3.1](https://ollama.com/library/llama3.1) 260 | - [LLaMA 3.1 Some Notes](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_llama-405b-is-here-and-it-comes-with-more-activity-7221533382025822208-K-Zm?utm_source=share&utm_medium=member_desktop) 261 | - [LLaMA 3.1 Model Finetunning](https://www.linkedin.com/posts/danielhanchen_google-colab-activity-7221621362417700867-y935/?utm_source=share&utm_medium=member_android) 262 | - [LLaMA 3.1 Detail Notes](https://www.linkedin.com/posts/sebastianraschka_yesterdays-llama-31-release-marked-a-big-activity-7221861717876645888-wz3H?utm_source=share&utm_medium=member_android) 263 | - [LLaMA 3.2 Detail Notes](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_llama-can-now-see-and-run-on-your-phone-activity-7244763879690354688-Iaan?utm_source=share&utm_medium=member_android) 264 | - [Mobile LLaMA 3.2](https://ai.meta.com/blog/llama-3-2-connect-2024-vision-edge-mobile-devices/) 265 | - [Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) 266 | - [How an online gifting site is using Llama to help protect customer privacy](https://ai.meta.com/blog/untukmu-built-with-llama/) [interesting] 267 | 268 | ### DeepSeek Models Related Links: 269 | - [DeepSeek-V3 Linkedin Post](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_yesterday-the-best-open-model-to-date-was-activity-7278313766679658498-6BCl?utm_source=share&utm_medium=member_desktop) 270 | - [Train your own R1 reasoning model with Unsloth (GRPO)](https://unsloth.ai/blog/r1-reasoning) 271 | 272 | ### Phi-3 Related Links: 273 | - [Phi-3 Linkedin Post1](https://www.linkedin.com/posts/sebastianraschka_microsoft-just-casually-shared-theirnew-activity-7188544168380510208-AdDG?utm_source=share&utm_medium=member_desktop) 274 | - [Phi-3 Linkedin Post2](https://www.linkedin.com/posts/julienchaumond_in-case-you-missed-it-earlier-this-week-activity-7189273186256003072-91B0?utm_source=share&utm_medium=member_desktop) 275 | 276 | ### Mistral & Mixtral Models Related Links: 277 | - [Mistral AI models](https://github.com/mistralai/mistral-src) 278 | - [Is Mistral's first model a good replacement for OpenAI?](https://blog.quivr.app/is-mistral-a-good-replacement-for-openai/) 279 | - [Mistral Mixture of Experts (MoE) Model](https://www.linkedin.com/posts/liorsinclair_big-news-mistral-just-released-an-open-source-activity-7139323993253228544-5coS?utm_source=share&utm_medium=member_desktop) 280 | - [Mixtral - a SOTA Mixture of Experts](https://huggingface.co/blog/mixtral) 281 | - [Mistraltrx](https://www.linkedin.com/posts/allen-roush-27721011b_cultrixmistraltrix-v1-hugging-face-activity-7149086757945298944-T7IA?utm_source=share&utm_medium=member_desktop) 282 | - [Nous-Hermes-Mixtral model](https://www.linkedin.com/posts/maxime-labonne_nousresearch-just-released-nous-hermes-activity-7152787405815566337-4aTY?utm_source=share&utm_medium=member_desktop) 283 | - [Mixtral in colab](https://github.com/dvmazur/mixtral-offloading/blob/master/notebooks/demo.ipynb) [Great] 284 | - [Brev.dev Notebooks: Fine-tuning mistral, mixtral, phi-2 and etc](https://github.com/brevdev/notebooks/tree/main) [**Excellent**] 285 | - [Optimized LLM inference api for mistral-7b using vllm and AWQ](https://lightning.ai/lightning-ai/studios/optimized-llm-inference-api-for-mistral-7b-using-vllm?view=public§ion=blogs) [**Excellent**] 286 | - [Run Mistral7b Quantized for free on any computer (CPU or GPU)](https://medium.com/artificial-corner/run-mistral7b-quantized-for-free-on-any-computer-2cadc18b45a2) [Interesting] 287 | - [Mixtral 8x22B a 176B MoE Model](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_new-open-model-from-mistral-ai-yesterday-activity-7183816273053523971-Vgse?utm_source=share&utm_medium=member_desktop) 288 | - [Mistral-7B-Instruct-v0.3](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_mistralaimistral-7b-instruct-v03-hugging-activity-7199103875348320256-lJ_A?utm_source=share&utm_medium=member_android) 289 | - [Codestral: A model fluent in 80+ programming languages](https://mistral.ai/news/codestral/) 290 | - [Mistral Finetune: the official repo and guide on how to fine-tune Mistral open-source models](https://github.com/mistralai/mistral-finetune) 291 | - [Mistral Large 2 Model](https://www.linkedin.com/posts/mistralai_large-enough-activity-7221915921622126593-JjHd?utm_source=share&utm_medium=member_desktop) 292 | - [Mistral Small 3](https://mistral.ai/news/mistral-small-3/) 293 | 294 | ### Yi Models: 295 | - [Yi Github](https://github.com/01-ai/Yi) 296 | - [Yi Website](https://01.ai/) 297 | - [Yi-VL-6B HuggingFace](https://huggingface.co/01-ai/Yi-VL-6B) 298 | 299 | ### Qwen Models: 300 | - [Introducing Qwen1.5 Blog Post](https://qwenlm.github.io/blog/qwen1.5/) 301 | - [Qwen1.5 Linkedin Post](https://www.linkedin.com/posts/andrew-iain-jardine_llm-opensource-llms-activity-7160905982523445248-_t5B?utm_source=share&utm_medium=member_desktop) 302 | - [Qwen1.5 HuggingFace](https://huggingface.co/collections/Qwen/qwen15-65c0a2f577b1ecb76d786524) 303 | - [Qwen2 HuggingFace](https://huggingface.co/docs/transformers/en/model_doc/qwen2) 304 | - [Qwen MoE Model](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_new-moe-alert-qwen15-moe-a27b-just-activity-7179144882668630016-i-l5?utm_source=share&utm_medium=member_android) 305 | - [Qwen2](https://github.com/QwenLM/Qwen2) 306 | - [Qwen 2.5 - Linkedin Post](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_9-new-multilingual-open-llms-released-qwen-activity-7242423229724676097-_9Ea?utm_source=share&utm_medium=member_desktop) 307 | - [Qwen 2.5 - Models](https://huggingface.co/collections/Qwen/qwen25-66e81a666513e518adb90d9e) 308 | 309 | ### DeepSeek Models: 310 | - [Huggingface DeepSeek R1 - Linkedin Post](https://www.linkedin.com/posts/qgallouedec_last-moments-of-closed-source-ai-hugging-activity-7288908822079852544-CDgF?utm_source=share&utm_medium=member_android) 311 | 312 | ### Gemma LLM Related Links (by Google): 313 | - [Gemma an open Gemini LLM released by Google! - Linkedin Post](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_welcome-gemma-googles-new-open-llm-activity-7166054332914741249-FY2D?utm_source=share&utm_medium=member_desktop) 314 | - [Gemma - another linkedin post](https://www.linkedin.com/posts/andrew-iain-jardine_opensource-llm-llms-activity-7166054662612226048-h0Ap?utm_source=share&utm_medium=member_desktop) 315 | - [Google's Gemma Detailed Notes](https://www.linkedin.com/posts/sebastianraschka_googles-gemma-has-been-the-topic-of-the-activity-7167160406480805888-PSeR?utm_source=share&utm_medium=member_desktop) 316 | - [Gemma usage via TRL](https://www.linkedin.com/posts/younes-belkada-b1a903145_new-release-from-google-gemma-a-state-of-the-art-activity-7166065899978870784-50To?utm_source=share&utm_medium=member_desktop) 317 | - [Gemma usage in Hugging Face via OpenAI SDK](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_yesterday-google-released-gemma-an-open-activity-7166484882917961730-uuFB?utm_source=share&utm_medium=member_desktop) 318 | - [Does Gemma overfit the Open LLM Leaderboard?](https://www.linkedin.com/posts/maxime-labonne_does-gemma-overfit-the-open-llm-leaderboard-activity-7166220798427402242-lJFm?utm_source=share&utm_medium=member_desktop) 319 | - [Zehpyr 7B Gemma](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_zehpyr-7b-gemma-releasedwe-are-excited-activity-7169373526641070080-rTLD?utm_source=share&utm_medium=member_desktop) 320 | - [Gemma 2](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_gemma-2-releasedgoogle-just-released-the-activity-7212108484920651776-BR8s?utm_source=share&utm_medium=member_desktop) 321 | - [Gemma2 Detailed Notes](https://www.linkedin.com/posts/sebastianraschka_whats-new-and-noteworthy-in-googles-newly-activity-7213528822384611329-sKv0?utm_source=share&utm_medium=member_desktop) 322 | - [Gemma 2-2b](https://huggingface.co/google/gemma-2-2b) 323 | 324 | ### Jamba (SSM-Transformer Model): 325 | - [AI21 Labs Jamba Model](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_jamba-released-ai21-labs-just-released-the-activity-7179121093482315776-xbmX?utm_source=share&utm_medium=member_desktop) 326 | - [Fine-tune jamba with TRL](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_yesterday-ai21-labs-released-jamba-the-first-activity-7179395299679858688-xiP9?utm_source=share&utm_medium=member_desktop) 327 | - [Fine-tune jamba code](https://www.linkedin.com/posts/maxime-labonne_jambatypus-v01-i-fine-tuned-a-jamba-activity-7181277758876962816-Z4zt?utm_source=share&utm_medium=member_desktop) 328 | 329 | ### 1-bit LLMs: 330 | - [1-bit LLMs (AlphaSignal Post)](https://www.linkedin.com/posts/liorsinclair_new-breakthrough-from-microsoft-1-bit-llms-activity-7168680301064384512-UeNv?utm_source=share&utm_medium=member_desktop) 331 | - [1-bit Quantization](https://www.linkedin.com/posts/a-roucher_%3F-%3F%3F%3F-%3F%3F%3F%3F%3F%3F%3F%3F%3F%3F%3F%3F-%3F%3F%3F%3F%3F%3F%3F%3F%3F%3F%3F%3F-activity-7168987208228540416-uhcm?utm_source=share&utm_medium=member_desktop) 332 | - [Some Notes about 1-bit LLMs (Their benefits and drawbacks)](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_the-era-of-1-bit-llms-what-does-that-mean-activity-7171533076668362753-Nl-F?utm_source=share&utm_medium=member_desktop) 333 | - [AutoBitnet (Train your 1.58-bit LLM based on LLama Architecture for free on Colab T4 GPU)](https://www.linkedin.com/posts/zaiinulabideen_autobitnet-train-your-158-bit-llm-based-activity-7182019658135326720-_qRp?utm_source=share&utm_medium=member_desktop) 334 | - [Llama2 7b in 1-bit precision](https://www.linkedin.com/posts/maxime-labonne_1-bit-quantization-activity-7179068277548032000-I8gR?utm_source=share&utm_medium=member_desktop) 335 | - [Microsoft 1-Bit LLM](https://github.com/microsoft/BitNet) 336 | 337 | ### Long Context Window LLMs (e.g., 100K Tokens LLMs): 338 | - [Claude LLM](https://www.linkedin.com/posts/itamar-g1_anthropic-openais-biggest-rivalry-just-activity-7063773334831775744-cQ4L/?utm_source=share&utm_medium=member_android) 339 | - [Some Notes about the 100K Claude LLM Model](https://www.linkedin.com/posts/sahar-mor_claude-a-gpt-competitor-from-anthropic-activity-7062811160168841216-z4u9/?utm_source=share&utm_medium=member_android) 340 | - [Anthropic's Claude-2](https://www.anthropic.com/index/claude-2) 341 | - [Claude-2, Anthropic's ChatGPT competitor](https://www.linkedin.com/posts/ugcPost-7084607703137857537-K9Ln?utm_source=share&utm_medium=member_desktop) 342 | - [Some Information about Claude 3](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_claude-3-is-here-anthropic-just-released-activity-7170424839529295872-Qp_S?utm_source=share&utm_medium=member_desktop) 343 | - [LongNet: Scaling Transformers to 1B Tokens](https://arxiv.org/abs/2307.02486) 344 | - [Lost in the Middle: How Language Models Use Long Contexts](https://arxiv.org/abs//2307.03172) 345 | - [Notes about How Language Models Use Long Contexts](https://www.linkedin.com/posts/sebastianraschka_llm-ai-machinelearning-activity-7083427280605089792-MS_N/?utm_source=share&utm_medium=member_android) 346 | - [Scaling LLaMA and GPTNeoX to >8k input context](https://www.linkedin.com/posts/gante_scaling-llama-and-gptneox-to-8k-input-context-activity-7085545793050320896-8OKi/?utm_source=share&utm_medium=member_android) 347 | - [Unofficial Claude-API](https://github.com/KoushikNavuluri/Claude-API) 348 | - [Claude Unofficial API](https://github.com/Explosion-Scratch/claude-unofficial-api) 349 | - [YARN & LongLlaMa](https://www.linkedin.com/posts/pramodith_generativeai-llm-gpt-activity-7104772654313656321-QC5D?utm_source=share&utm_medium=member_desktop) 350 | - [YaRN: Efficient Context Window Extension of LLMs](https://github.com/jquesnelle/yarn) 351 | - [LLMs get lost when the context becomes too long: Lost in the Middle: How Language Models Use Long Contexts](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_are-vector-databases-here-to-stay-yes-activity-7085908435686285312-QVfB?utm_source=share&utm_medium=member_desktop) [**Very Important**] 352 | - [LongLoRA: Efficient Fine-tuning of Long-Context LLMs](https://www.linkedin.com/posts/omarsar_longlora-efficient-fine-tuning-of-long-context-activity-7111000280615325699-SVEE?utm_source=share&utm_medium=member_desktop) 353 | - [LongLoRA: Efficient Fine-tuning of Long-Context LLMs (another post)](https://www.linkedin.com/posts/haotian-tang_expanding-the-context-size-of-large-language-activity-7110806911775641600-nShH?utm_source=share&utm_medium=member_desktop) 354 | - [Efficient Streaming LLMs with Attention Sinks for infinite-length inputs](https://github.com/mit-han-lab/streaming-llm) 355 | - [MemGPT: Teaching LLMs memory management for unbounded context](https://github.com/cpacker/MemGPT) 356 | - [LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs](https://github.com/THUDM/LongWriter) [Interesting] 357 | - [Llmlingua Prompt Compress](https://www.linkedin.com/posts/sahar-mor_microsoft-recently-published-a-new-technique-activity-7151596182379597825-7ego?utm_source=share&utm_medium=member_desktop) [Interesting] 358 | 359 | ### Small Language Models (SLMs): 360 | - [Microsoft Phi-2 Model (with 2.7B Parameters)](https://www.microsoft.com/en-us/research/blog/phi-2-the-surprising-power-of-small-language-models/) 361 | - [Can "small" finetuned LLMs with less than 2B parameters outperform larger openly available LLMs (Mixtral, Llama 2 Chat) and proprietary LLMs (ChatGPT)?](https://www.linkedin.com/posts/sebastianraschka_can-small-finetuned-llms-with-less-than-activity-7162082013674500096-FuYV?utm_source=share&utm_medium=member_desktop) 362 | - [Smol LM](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_in-just-a-few-years-the-majority-of-ai-usage-activity-7219027139352801281-aMYy?utm_source=share&utm_medium=member_desktop) 363 | - [Hymba Small LM](https://www.linkedin.com/posts/pavlo-molchanov-08738a63_excited-to-share-a-new-efficient-small-language-activity-7265581836582424576-4Mqp?utm_source=share&utm_medium=member_desktop) 364 | 365 | ### Frameworks for Training & Using Large Language Models (LLMs): 366 | - [ColossalAI: Library for LLMs](https://github.com/hpcaitech/ColossalAI) 367 | - [LangChain: Library for Building applications with LLMs](https://github.com/hwchase17/langchain) 368 | - [LangChain Chat](https://github.com/hwchase17/chat-langchain) 369 | - [LangChain Crash Course](https://www.youtube.com/watch?v=LbT1yp6quS8) 370 | - [LangChain 101](https://www.linkedin.com/posts/munjal-patel_llm-chatgpt-machinelearning-activity-7049757220300800000-hH7I/?utm_source=share&utm_medium=member_android) 371 | - [LangChain Resources](https://www.linkedin.com/posts/sonali-pattnaik_generativeai-ai-activity-7063160223967973376-3K0P/?utm_source=share&utm_medium=member_android) 372 | - [LangChain & Vector Databases in Production Course](https://learn.activeloop.ai/courses/langchain) 373 | - [Building LLM Powered Apps via LangChain Course](https://www.wandb.courses/courses/building-llm-powered-apps) 374 | - [OpenFlamingo](https://github.com/mlfoundations/open_flamingo) 375 | - [Deepset Haystack Framework](https://github.com/deepset-ai/haystack) 376 | - [LMQL: A query language for programming LLMs](https://github.com/eth-sri/lmql) 377 | - [LLM Training Frameworks List](https://www.linkedin.com/posts/aboniasojasingarayar_llm-gpt3-framework-activity-7047449940192591872-3VYc/?utm_source=share&utm_medium=member_android) 378 | - [NeMo Guardrails](https://github.com/NVIDIA/NeMo-Guardrails) 379 | - [Lamini: The LLM engine for rapidly customizing models](https://github.com/lamini-ai/lamini) 380 | - [Scikit-LLM: Sklearn Meets Large Language Models](https://github.com/iryna-kondr/scikit-llm) 381 | - [Chainlit](https://github.com/Chainlit/chainlit) 382 | - [ChatUI](https://github.com/alibaba/ChatUI) 383 | - [Streamlit-Chat](https://github.com/AI-Yash/st-chat) 384 | - [Gradio: Creating a Streaming chatbot fast](https://www.gradio.app/guides/creating-a-chatbot-fast#streaming-chatbots) 385 | - [Streamlit-Weaviate Connection: provides a custom streamlit connection to query data from weaviate](https://github.com/weaviate/st-weaviate-connection/tree/main) 386 | - [LangKit: an open-source text metrics toolkit for monitoring language models](https://github.com/whylabs/langkit) 387 | - [HuggingFace Transformers Agents](https://huggingface.co/docs/transformers/transformers_agents) 388 | - [privateGPT: Ask questions to your documents using the power of LLMs](https://github.com/imartinez/privateGPT) 389 | - [Spacy LLM](https://github.com/explosion/spacy-llm) 390 | - [Lit-GPT](https://github.com/Lightning-AI/lit-gpt) 391 | - [Zero to LitGPT Tutorial: Getting Started with Pretraining, Finetuning, and Using LLMs](https://github.com/Lightning-AI/litgpt/blob/main/tutorials/0_to_litgpt.md) [Great] 392 | - [GPTCache: A Library for Creating Semantic Cache for LLM Queries](https://github.com/zilliztech/GPTCache/tree/main) 393 | - [AutoTrain-Advanced](https://github.com/huggingface/autotrain-advanced) 394 | - [Monster API: API for using & fine-tuning LLMs](https://monsterapi.ai/) 395 | - [AnythingLLM: A full-stack personalized AI assistant](https://github.com/Mintplex-Labs/anything-llm) 396 | - [EasyLLM: helpful tools and methods for working with LLMs](https://github.com/philschmid/easyllm) 397 | - [gpt-llm-trainer: input a description of your task, and fine-tune a LLaMA 2 model for you](https://github.com/mshumer/gpt-llm-trainer) 398 | - [Embedchain: a framework to easily create LLM powered bots](https://github.com/embedchain/embedchain) 399 | - [PandasAI](https://github.com/gventuri/pandas-ai) [It is not related strictly in this section, but it is interesting] 400 | - [GPT Engineer: Specify what you want it to build, the AI asks for clarification, and then builds it](https://github.com/AntonOsika/gpt-engineer) 401 | - [Ludwig: a low-code framework for building custom AI models like LLMs](https://github.com/ludwig-ai/ludwig) 402 | - [open-interpreter](https://github.com/KillianLucas/open-interpreter) 403 | - [kani: is a lightweight and highly hackable framework for chat-based language models with tool usage/function calling](https://github.com/zhudotexe/kani) 404 | - [Kani colab samples](https://colab.research.google.com/github/zhudotexe/kani/blob/main/examples/colab_examples.ipynb) 405 | - [Kani Linkedin Post](https://www.linkedin.com/posts/chris-callison-burch-40bb87b7_my-phd-students-have-build-a-really-great-activity-7110728026971115520-T16F?utm_source=share&utm_medium=member_desktop) 406 | - [Argilla: the open-source data curation platform for LLMs](https://github.com/argilla-io/argilla) 407 | - [LiteLLM: Call all LLM APIs using the OpenAI format](https://github.com/BerriAI/litellm) 408 | - [LLM Finetuning with PEFT](https://github.com/ashishpatel26/LLM-Finetuning) 409 | - [ChatGPT-AutoExpert: Supercharged Custom Instructions for ChatGPT](https://github.com/spdustin/ChatGPT-AutoExpert) 410 | - [PyTorch thunder (pytorch compiler for speed up training of LLMs) - Linkedin Post](https://www.linkedin.com/posts/sebastianraschka_we-just-open-sourced-thunder-a-new-compiler-activity-7176571765639245824-srIZ?utm_source=share&utm_medium=member_desktop) 411 | - [PyTorch Lightning Thunder](https://github.com/Lightning-AI/lightning-thunder) 412 | - [unsloth library: 2-5X faster 70% less memory QLoRA & LoRA finetuning](https://github.com/unslothai/unsloth) [**Great for fine-tuning LLMs**] 413 | - [TorchTune: A Native-PyTorch Library for LLM Fine-tuning](https://github.com/pytorch/torchtune) 414 | 415 | ### Notes and Codes for Training and fine-tuning LLMs: 416 | - [LLM Finetuning with PEFT Colab Notebooks](https://github.com/ashishpatel26/LLM-Finetuning) 417 | - [Self Instruct TRL for LLMs](https://github.com/yizhongw/self-instruct) 418 | - [Self Instruct TRL for LLMs - Link2](https://huggingface.co/docs/trl/sft_trainer) 419 | - [How to Fine-Tune LLMs in 2024 with Hugging Face](https://www.philschmid.de/fine-tune-llms-in-2024-with-trl) 420 | - [How to fine-tune open LLMs in 2025 with Hugging Face](https://www.philschmid.de/fine-tune-llms-in-2025) 421 | - [Fine tune LLMs in your own hardware via PyTorch team (great)](https://pytorch.org/blog/finetune-llms/?utm_content=278057355&utm_medium=social&utm_source=linkedin&hss_channel=lcp-78618366) 422 | - [RLHF in 2024 with DPO & Hugging Face](https://www.philschmid.de/dpo-align-llms-in-2024-with-trl) 423 | - [A little guide to building Large Language Models in 2024 (PPT by HuggingFace Team)](https://docs.google.com/presentation/d/1IkzESdOwdmwvPxIELYJi8--K3EZ98_cL6c5ZcLKSyVg/edit?usp=sharing) [**Great**] 424 | - [Video Link1 of A little guide to building Large Language Models in 2024 (PPT by HuggingFace Team)](https://www.linkedin.com/posts/thom-wolf_75min-talk-i-finally-recorded-this-lecture-activity-7179106246505967617-0nzC?utm_source=share&utm_medium=member_desktop) 425 | - [Video Link2 of A little guide to building Large Language Models in 2024 (PPT by HuggingFace Team)](https://www.youtube.com/watch?v=2-SPH9hIKT8) 426 | - [Understanding the instruction fine-tuning process in LLMs](https://www.linkedin.com/posts/sebastianraschka_if-you-are-looking-for-a-resource-to-understand-activity-7208093607122145280-6wFF?utm_source=share&utm_medium=member_desktop) 427 | - [Top 5 Tips and Tricks for LLM Fine-Tuning and Inference from Intel Experts](https://www.intel.com/content/www/us/en/developer/articles/technical/top-tricks-for-llm-fine-tuning-and-inference.html) 428 | 429 | ### Reflection-Tuning of LLMs: 430 | - [Reflection-Tuning of LLMs](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_mindblowing-a-70b-open-meta-llama-3-better-activity-7237712642339926016-Cfm6?utm_source=share&utm_medium=member_desktop) 431 | 432 | ### Memory Layer for LLMs: 433 | - [Memory layer for LLMs](https://www.linkedin.com/posts/liorsinclair_mem0-gained-20000-stars-on-github-in-30-activity-7237475167822585857-4Jbu?utm_source=share&utm_medium=member_desktop) 434 | - [Memory layer for LLMs - GitHub Repo](https://github.com/mem0ai/mem0) 435 | 436 | ### LLMs for Coding: 437 | - [CodeGen](https://github.com/salesforce/CodeGen) 438 | - [Code Llama](https://github.com/facebookresearch/codellama) 439 | - [Code Llama Notes](https://www.linkedin.com/posts/aleksagordic_nice-meta-ai-just-announced-code-llama-activity-7100559934764810240-Un2i/?utm_source=share&utm_medium=member_android) 440 | 441 | ### LLMs as Front-End Engineers: 442 | - [Design2Code: How Far Are We From Automating Front-End Engineering?](https://arxiv.org/abs/2403.03163) 443 | - [Llama Coder: Can generate full React apps](https://llamacoder.together.ai/) 444 | 445 | ### LLMs Courses & Tutorials: 446 | - [LLM Bootcamp - Spring 2023](https://fullstackdeeplearning.com/llm-bootcamp/spring-2023/) 447 | - [LLM University](https://docs.cohere.com/docs/llmu) 448 | - [List of LLM Courses](https://www.linkedin.com/posts/srijankr_ai-llm-activity-7080929772523966464-Le4u/?utm_source=share&utm_medium=member_android) 449 | - [Anti-hype LLM reading list](https://gist.github.com/veekaybee/be375ab33085102f9027853128dc5f0e) 450 | - [Microsoft Generative AI Course](https://github.com/microsoft/generative-ai-for-beginners) 451 | - [Google and Kaggle five-day generative AI course](https://blog.google/technology/developers/google-kaggle-genai-intensive/) [Good] 452 | - [Best Resources for learning to work with LLMs](https://www.linkedin.com/posts/whats-ai_github-louisfb01start-llms-a-complete-activity-7133590058229456896-WEf0?utm_source=share&utm_medium=member_desktop) 453 | - [Start with Large Language Models (LLMs) - Become an expert for free!](https://github.com/louisfb01/start-llms) [Interesting] 454 | - [Intro to LLMs: Andrej Karpathy 1 Hour Lecture](https://www.youtube.com/watch?v=zjkBMFhNj_g) 455 | - [LLM Course](https://github.com/mlabonne/llm-course) [**good**] 456 | - [LLM Course in ChatGPT Plus](https://www.linkedin.com/posts/maria-vechtomova_llm-gpt-activity-7160567161856360448-IFjd?utm_source=share&utm_medium=member_desktop) 457 | - [Build a Large Language Model (From Scratch) great Course and Book Tutorial](https://github.com/rasbt/LLMs-from-scratch) [**Great**] 458 | - [Learning Resources about LLMs](https://www.linkedin.com/posts/pauliusztin_machinelearning-mlops-datascience-activity-7135530424767819777-ui-5?utm_source=share&utm_medium=member_desktop) 459 | - [The Transformer Layer by Layer Course](https://mlbootcamp.ai/course.html?guid=d105240a-94e1-405b-be80-60056659c24c) 460 | - [The Transformer Layer by Layer Course: Linkedin](https://www.linkedin.com/posts/juan-olano-b9a330112_artificialintelligence-transformers-onlinelearning-activity-7137158122715897856-cneV?utm_source=share&utm_medium=member_desktop) 461 | - [Hands-on LLMs Course](https://github.com/iusztinpaul/hands-on-llms) 462 | - [Direct Preference Optimization (DPO) Method for LLMs Tutorial](https://huggingface.co/blog/pref-tuning) 463 | - [CS25: Transformers United V3 Courses - Autumn 2023](https://web.stanford.edu/class/cs25/) 464 | - [CS336: Language Modeling from Scratch](https://stanford-cs336.github.io/spring2024/) 465 | - [Visual and Animated Lecture about LLMs and Transformers and Deep Learning](https://www.youtube.com/playlist?list=PLZHQObOWTQDNU6R1_67000Dx_ZCJB-3pi) 466 | - [LLMs Roadmap](https://www.linkedin.com/posts/ba%C5%9Fak-tu%C4%9F%C3%A7e-eskili-61511b58_nlp-llms-gpt3-activity-7168168071356997632-V8yL?utm_source=share&utm_medium=member_desktop) [Great] 467 | - [Brev.dev Notebooks: Fine-tuning mistral, mixtral, phi-2 and etc](https://github.com/brevdev/notebooks/tree/main) [**Excellent**] 468 | - [LLM Summer School](https://www.linkedin.com/posts/sebastianraschka_a-suggestion-for-an-effective-11-step-llm-activity-7195778889384693762-2TB_?utm_source=share&utm_medium=member_android) 469 | - [LLM Engineer's Handbook](https://www.linkedin.com/posts/maxime-labonne_super-proud-to-announce-my-new-book-the-activity-7219253497559425024-IVkc?utm_source=share&utm_medium=member_desktop) 470 | - [LLM Twin Course: Building Your Production-Ready AI Replica](https://github.com/decodingml/llm-twin-course) 471 | - [Hands-On Large Language Models Book](https://www.linkedin.com/posts/jalammar_our-newly-released-llm-book-hands-on-large-activity-7242207044533948417-_i2R?utm_source=share&utm_medium=member_desktop) 472 | - [Foundations of LLMs Book](https://arxiv.org/abs/2501.09223) 473 | 474 | ### LLMs Ranking: 475 | - [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) 476 | - [Chatbot Arena Leaderboard](https://lmsys.org/blog/2023-05-10-leaderboard/) 477 | - [AlpacaEval Leaderboard](https://tatsu-lab.github.io/alpaca_eval/) 478 | - [CanAiCode Leaderboard](https://huggingface.co/spaces/mike-ravkine/can-ai-code-results) 479 | - [Small LLMs Performance Ranking](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_how-big-do-llms-need-to-be-able-to-reason-activity-7134108036473741312-2jxI?utm_source=share&utm_medium=member_desktop) 480 | - [Chatbot Arena: Benchmarking LLMs in the Wild](https://huggingface.co/spaces/lmsys/chatbot-arena) [**Great**] 481 | - [Chatbot Arena Leaderboard](https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard) 482 | - [AI2 WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild](https://huggingface.co/spaces/allenai/WildBench) [**Great**] 483 | - [AI2 WildBench Linkedin Post](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_new-evaluation-benchmark-leaderboard-by-activity-7171853629325316096-67sr?utm_source=share&utm_medium=member_desktop) 484 | - [Persian LLM Leaderboard (via Part AI)](https://huggingface.co/spaces/PartAI/persian-llm-leaderboard) 485 | 486 | ### Building NLP Applications Powered by LLMs (Different Methods for Augmenting Knowledge to LLMs (or Retrieval-Augmented Generation (RAG) applications)): 487 | - [Ask a Book Questions with LangChain OpenAI](https://bennycheung.github.io/ask-a-book-questions-with-langchain-openai) [Great] 488 | - [OpenAI Web QA Embeddings](https://platform.openai.com/docs/tutorials/web-qa-embeddings) 489 | - [Deepset Haystack Framework](https://github.com/deepset-ai/haystack) 490 | - [Stanford Retrieval-based NLP](https://ai.stanford.edu/blog/retrieval-based-NLP/) 491 | - [Hypothetical Document Embeddings (HyDE)](https://www.linkedin.com/posts/activity-7048838677438861312-8MFD/?utm_source=share&utm_medium=member_android) 492 | - [ChatDB: Augmenting LLMs with Databases](https://chatdatabase.github.io/) 493 | - [ChatNode](https://www.chatnode.ai/) 494 | - [Emerging Architectures for LLM Applications](https://a16z.com/2023/06/20/emerging-architectures-for-llm-applications/) 495 | - [Evaluation framework for your Retrieval Augmented Generation (RAG) pipelines](https://github.com/explodinggradients/ragas) 496 | - [Fine tuning vs. RAG for LLMs](https://www.linkedin.com/posts/alexander-ratner-038ba239_lots-of-debate-on-fine-tuning-vs-rag-for-activity-7103836027957506048-AjoJ?utm_source=share&utm_medium=member_desktop) 497 | - [Building RAG-based LLM Applications for Production (Part 1)](https://www.anyscale.com/blog/a-comprehensive-guide-for-building-rag-based-llm-applications-part-1) [Good] 498 | - [Verba: The Golden RAGtriever, user-friendly interface for Retrieval-Augmented Generation (RAG) applications](https://github.com/weaviate/Verba) 499 | - [DocsGPT: GPT-powered chat for documentation, chat with your documents](https://github.com/arc53/DocsGPT) 500 | - [RAFT: Retrieval Augmented Fine Tuning - Post1](https://www.linkedin.com/posts/pascalbiese_raft-the-best-of-rag-and-fine-tuning-combined-activity-7175089937036283904-ltQI?utm_source=share&utm_medium=member_desktop) 501 | - [RAFT: Retrieval Augmented Fine Tuning - Post2](https://www.linkedin.com/posts/tianjun-zhang-333bb2126_raft-a-new-way-to-teach-llms-to-be-better-activity-7174525633291587584-CO-h?utm_source=share&utm_medium=member_desktop) 502 | - [RAFT: Retrieval Augmented Fine Tuning - Microsoft Blog](https://techcommunity.microsoft.com/t5/ai-ai-platform-blog/raft-a-new-way-to-teach-llms-to-be-better-at-rag/ba-p/4084674) 503 | - [RAFT: Retrieval Augmented Fine Tuning - Berkeley Blog](https://gorilla.cs.berkeley.edu/blogs/9_raft.html) 504 | - [RAFT Code](https://github.com/ShishirPatil/gorilla/tree/main/raft) 505 | - [Long context LLMs vs RAG](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_how-good-are-llms-in-a-long-context-and-activity-7214185350959689728-cnfp?utm_source=share&utm_medium=member_android) [Interesting] 506 | - [RAGFlow: an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding](https://github.com/infiniflow/ragflow) 507 | - [Two Step RAG: Speculative RAG: Enhancing retrieval augmented generation through drafting](https://research.google/blog/speculative-rag-enhancing-retrieval-augmented-generation-through-drafting/) 508 | - [Exploring Multimodal RAG with LlamaIndex and GPT-4 or the New Anthropic Sonnet Model](https://levelup.gitconnected.com/exploring-multimodal-rag-with-llamaindex-and-gpt-4-or-the-new-anthropic-sonnet-model-96705c877dbb) 509 | - [PaperQA2: High accuracy RAG for answering questions from scientific documents with citations](https://github.com/Future-House/paper-qa) 510 | - [Sophisticated Controllable Agent for Complex RAG Tasks](https://github.com/NirDiamant/Controllable-RAG-Agent) 511 | - [Anthropic's Cluade Introducing Contextual Retrieval RAG](https://www.anthropic.com/news/contextual-retrieval) 512 | - [Docling: Get your docs ready for gen AI](https://github.com/DS4SD/docling) 513 | - [Lecture of RAG and Prompt Engineering](https://www.linkedin.com/posts/tom-yeh_i-just-edited-my-lecture-beginners-guide-activity-7284242137091620864-6MBy?utm_source=share&utm_medium=member_desktop) 514 | - [Recent RAG Research from Google](https://www.linkedin.com/posts/jihoo-kim_rag-research-from-google-2024-ugcPost-7266537405904498689-wrac?utm_source=share&utm_medium=member_android) 515 | - [zoekt: Fast trigram based code search --> great tool for RAG of codes](https://github.com/sourcegraph/zoekt) [**important**] 516 | 517 | ### Graph RAG & Its Related Data Bases: 518 | - [ArangoDB: The Most Complete And Scalable Platform For Graph-Powered GenAI](https://arangodb.com/) 519 | - [Microsoft GraphRAG](https://microsoft.github.io/graphrag/) 520 | - [llamaindex Graph RAG](https://docs.llamaindex.ai/en/stable/examples/query_engine/knowledge_graph_rag_query_engine/) 521 | - [Gephi: The Open Graph Viz Platform](https://gephi.org/) 522 | - [JanusGraph: is a scalable graph database optimized for storing and querying graphs](https://janusgraph.org/) 523 | - [cayley: Open Source Graph Data Base](https://cayley.io/) 524 | - [Retrieval-Augmented Generation with Knowledge Graphs for Customer Service Question Answering (Paper)](https://arxiv.org/abs/2404.17723) 525 | - [The GraphRAG Manifesto: Adding Knowledge to GenAI](https://neo4j.com/blog/graphrag-manifesto/) 526 | - [Neo4j for GenAI](https://neo4j.com/generativeai/) 527 | 528 | ### Cache-Augmented Generation (CAG): 529 | - [Cache-Augmented Generation (CAG) - Linkedin Post1](https://www.linkedin.com/posts/maryammiradi_dont-do-rag-cag-is-40x-faster-than-activity-7281655697086287872-c35Q?utm_source=share&utm_medium=member_desktop) 530 | - [Cache-Augmented Generation (CAG) - Linkedin Post2](https://www.linkedin.com/posts/bhavishya-pandit_rag-vs-cag-activity-7282615153852862464-ES23?utm_source=share&utm_medium=member_desktop) 531 | - [Cache-Augmented Generation (CAG) - Linkedin Post3](https://www.linkedin.com/posts/francoisvanderseypen_dont-do-rag-when-cache-augmented-generation-activity-7279725990342193152-8P82?utm_source=share&utm_medium=member_desktop) 532 | 533 | ### Vector Database Libraries: 534 | - [weaviate](https://weaviate.io/) 535 | - [weaviate GitHub](https://github.com/weaviate/weaviate) 536 | - [chroma](https://github.com/chroma-core/chroma) 537 | - [Qdrant: Vector Database for AI Applications](https://github.com/qdrant/qdrant) 538 | - [pinecone](https://www.pinecone.io/) 539 | - [rektor-db](https://github.com/codediodeio/rektor-db) 540 | - [pgvector](https://github.com/pgvector/pgvector) 541 | - [LlamaIndex: comprehensive toolkit to perform data augmentation for LLMs](https://github.com/jerryjliu/llama_index) 542 | - [jina-ai VectorDB](https://github.com/jina-ai/vectordb) 543 | - [sqlite-vec: A vector search SQLite extension](https://github.com/asg017/sqlite-vec) 544 | 545 | ### Great Embedding Models for Search (for Augmenting External Knowledge into ChatBot Vector DB) [Retrieval Augmented Generation (RAG)]: 546 | - [Massive Text Embedding Benchmark (MTEB) Leaderboard](https://huggingface.co/spaces/mteb/leaderboard) 547 | - [Word and sentence embeddings is how LLMs understand text](https://www.linkedin.com/posts/sahar-mor_word-and-sentence-embeddings-is-how-llms-activity-7105921473978015744-R0Nm?utm_source=share&utm_medium=member_desktop) 548 | - [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding) 549 | - [E5 embedding vs OpenAI Ada](https://www.linkedin.com/posts/andrew-iain-jardine_hosting-a-text-embedding-model-that-is-better-activity-7106338837479510016-zvBW?utm_source=share&utm_medium=member_desktop) 550 | - [M2-BERT-80M-32k-Retrieval](https://huggingface.co/togethercomputer/m2-bert-80M-32k-retrieval) 551 | - [Embedding Quantization - Post1](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_introducing-embedding-quantization-a-new-activity-7176971093646159872-hp9z?utm_source=share&utm_medium=member_desktop) 552 | - [Embedding Quantization - Post2](https://www.linkedin.com/posts/tomaarsen_binary-and-scalar-embedding-quantization-activity-7176966403332132864-lJzH?utm_source=share&utm_medium=member_desktop) 553 | - [Embedding Quantization - HuggingFace Blog Post](https://huggingface.co/blog/embedding-quantization) 554 | - [Quantization Fundamentals with Hugging Face Course](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_quantization-fundamentals-with-hugging-face-activity-7186335433843167232-sKV2?utm_source=share&utm_medium=member_desktop) 555 | - [Is Cosine-Similarity of Embeddings Really About Similarity?](https://www.linkedin.com/posts/alphasignal_is-cosine-similarity-of-embeddings-really-activity-7175543620651880449-ZoKw?utm_source=share&utm_medium=member_desktop) 556 | - [LLM2Vec](https://www.linkedin.com/posts/zaiinulabideen_lazy-llm2vec-convert-your-favorite-llm-activity-7193618083448553472-_Q2e?utm_source=share&utm_medium=member_desktop) [**Great**] 557 | - [Fine tuning embedding models for RAG (Linkedin post)](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_fine-tune-embedding-models-for-retrieval-activity-7203760204579028992-g7eW?utm_source=share&utm_medium=member_desktop) 558 | - [Fine tuning embedding models for RAG (Original Post)](https://www.philschmid.de/fine-tune-embedding-model-for-rag) 559 | - [`all-MiniLM-L6-v2` --> Sentence-Transformers Model for Embedding](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) 560 | - [Learn How to Fine-tuning Embedding Models Course](https://marqo.ai/courses/fine-tuning-embedding-models) [**Great**] 561 | - [LLMs Embedding Course - Link1](https://github.com/anishiisc/Build_LLM_from_Scratch/tree/main) 562 | - [LLMs Embedding Course - Link2](https://www.linkedin.com/posts/ugcPost-7228118123390902272-oVu4/?utm_source=share&utm_medium=member_android) 563 | - [txtai: All-in-one embeddings database](https://github.com/neuml/txtai) 564 | - [NVIDIA NV-emb-2 embeddings](https://www.linkedin.com/posts/tunguz_ok-nvidia-nv-emb-2-embeddings-are-really-activity-7262862383885213696-MWVv?utm_source=share&utm_medium=member_desktop) 565 | - [jina-embeddings-v3: Multilingual Embeddings With Task LoRA](https://huggingface.co/papers/2409.10173) 566 | - [ModernBert: Linkedin Post1](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_modernbert-bert-revisited-in-the-age-of-activity-7275551060302131201-dr3c?utm_source=share&utm_medium=member_desktop) 567 | - [ModernBert: Linkedin Post2](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_want-to-replace-bert-in-2025-the-time-has-activity-7277616689859444737-iRUe?utm_source=share&utm_medium=member_desktop) 568 | - [Nomic-embed-text-v2-moe model](https://huggingface.co/nomic-ai/nomic-embed-text-v2-moe) 569 | - [Nomic Embed Text V2: An Open Source, Multilingual, Mixture-of-Experts Embedding Model (Blog Post)](https://www.nomic.ai/blog/posts/nomic-embed-text-v2) 570 | - [Gemeni models for text embedding (original link)](https://developers.googleblog.com/en/gemini-embedding-text-model-now-available-gemini-api/) 571 | - [Gemeni models for text embedding (useful linkedin post)](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_gemini-models-for-embeddings-yes-google-activity-7303840326933176320-Tg0C?utm_source=share&utm_medium=member_android&rcm=ACoAAAgksdYBFu3_vG0bwXWdh93rSqV1J1ghMP4) 572 | 573 | ### Prevent Hallucinations from LLMs & Controling their outputs: 574 | - [Deep Dive Into LLM Hallucinations Across Generative Tasks](https://www.rungalileo.io/blog/deep-dive-into-llm-hallucinations-across-generative-tasks) 575 | - [Controlled Generation Tools](https://www.linkedin.com/posts/pascalbiese_genai-llms-opensource-activity-7097185067885576192-Uv8Z/?utm_source=share&utm_medium=member_android) 576 | - [Guidance: Controlling LLMs](https://github.com/guidance-ai/guidance) 577 | - [NeMo Guardrails](https://github.com/NVIDIA/NeMo-Guardrails) 578 | - [Minimising Hallucinations in LLM Applications: NeMo Guradrails Video Tutorial](https://www.linkedin.com/posts/sanyambhutani_minimising-hallucinations-in-llm-applications-activity-7104810583304077312-w983?utm_source=share&utm_medium=member_desktop) 579 | - [Mitigate Hallucination in LLMs](https://www.linkedin.com/posts/vinija_mitigate-hallucination-in-llms-as-activity-7114468991330390016-O0BZ?utm_source=share&utm_medium=member_desktop) 580 | - [LLMs Hallucinations Benchmark](https://www.linkedin.com/posts/drjimfan_please-see-update-below-a-recent-llm-hallucination-activity-7130230516246593536-mxAY?utm_source=share&utm_medium=member_desktop) 581 | - [Mitigating LLM Hallucinations: a multifaceted approach](https://amatriain.net/blog/hallucinations) [Great] 582 | 583 | ### Training & Using Large Language Models (LLMs) on Low Resource Machines: 584 | - [Cramming: Training a Language Model on a Single GPU in One Day](https://github.com/jonasgeiping/cramming) [**Great**] 585 | - [Fine-tuning 20B LLMs with RLHF on a 24GB consumer GPU](https://huggingface.co/blog/trl-peft) [**Great**] 586 | - [PEFT: State-of-the-art Parameter-Efficient Fine-Tuning](https://github.com/huggingface/peft) [**Great**] 587 | - [PEFT: Parameter-Efficient Fine-Tuning of Billion-Scale Models on Low-Resource Hardware](https://huggingface.co/blog/peft) [**Great**] 588 | - [Introduction to 8-bit Matrix Multiplication for transformers at scale using Hugging Face Transformers, Accelerate and bitsandbytes](https://huggingface.co/blog/hf-bitsandbytes-integration) 589 | - [bitsandbytes: 8-bit CUDA functions for PyTorch](https://github.com/TimDettmers/bitsandbytes) 590 | - [Alpaca-LoRA: Low-Rank LLaMA Instruct-Tuning on consumer hardware](https://github.com/tloen/alpaca-lora) [Great] 591 | - [LLaMA & Alpaca Tutorial: “ChatGPT” On Your Local Computer](https://medium.com/@martin-thissen/llama-alpaca-chatgpt-on-your-local-computer-tutorial-17adda704c23) 592 | - [Dalai: The simplest way to run LLaMA on your local machine](https://github.com/cocktailpeanut/dalai) 593 | - [pyllama](https://github.com/juncongmoo/pyllama) 594 | - [Alpaca-LoRA-Serve](https://github.com/deep-diver/Alpaca-LoRA-Serve) 595 | - [llama.cpp: Port of Facebook's LLaMA model in C/C++](https://github.com/ggerganov/llama.cpp) 596 | - [alpaca.cpp](https://github.com/antimatter15/alpaca.cpp) 597 | - [SparseGPT: Remove 100 Billion Parameters of LLMs](https://neuralmagic.com/blog/sparsegpt-remove-100-billion-parameters-for-free/) 598 | - [xFormers: Toolbox to Accelerate Research on Transformers](https://github.com/facebookresearch/xformers) 599 | - [LLaMA-Adapter: Efficient Fine-tuning of LLaMA (Fine-tuning LLaMA to follow instructions within 1 Hour and 1.2M Parameters)](https://github.com/ZrrSkywalker/LLaMA-Adapter) 600 | - [GPT4All](https://github.com/nomic-ai/gpt4all) [Great] 601 | - [Vicuna web page](https://vicuna.lmsys.org/) [Great] 602 | - [Vicuna GitHub: FastChat](https://github.com/lm-sys/FastChat) 603 | - [PetGPT](https://github.com/maziarraissi/PetGPT) 604 | - [GPT-4-LLM](https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM) 605 | - [baize Chatbot](https://github.com/project-baize/baize-chatbot) 606 | - [Koala](https://github.com/young-geng/EasyLM#koala) 607 | - [Gorilla: An API store for LLMs](https://github.com/ShishirPatil/gorilla) 608 | - [Lit-LLaMA](https://github.com/Lightning-AI/lit-llama) 609 | - [Auto-GPT](https://github.com/Torantulino/Auto-GPT) 610 | - [xTuring](https://github.com/stochasticai/xTuring) 611 | - [GPTCache](https://github.com/zilliztech/gptcache) 612 | - [Dolly-v2-12B](https://huggingface.co/databricks/dolly-v2-12b) 613 | - [Web LLM](https://github.com/mlc-ai/web-llm) 614 | - [P-tuning v2](https://github.com/THUDM/P-tuning-v2) 615 | - [QLoRA: Efficient Finetuning of Quantized LLMs](https://github.com/artidoro/qlora) 616 | - [AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration](https://github.com/mit-han-lab/llm-awq) 617 | - [GPTQ Quantization Method in Transformers](https://www.linkedin.com/posts/marc-sun_opensource-llm-quantization-activity-7100102215582797824-td7E?utm_source=share&utm_medium=member_desktop) 618 | - [Optimize open LLMs using GPTQ and Hugging Face Optimum](https://www.linkedin.com/feed/update/urn:li:activity:7103049470908485632/?utm_source=share&utm_medium=member_android) 619 | - [GPTQ vs. bitsandbytes (BNB)](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_quantization-makes-fine-tuning-and-deploying-activity-7104480375841636352-_dgY?utm_source=share&utm_medium=member_desktop) 620 | - [BNB Blog: Making LLMs even more accessible with bitsandbytes, 4-bit quantization and QLoRA](https://huggingface.co/blog/4bit-transformers-bitsandbytes) 621 | - [GPTQ Blog: Making LLMs lighter with AutoGPTQ and transformers](https://huggingface.co/blog/gptq-integration) 622 | - [TensorRT-LLM](https://www.linkedin.com/posts/tunguz_llm-h100-languagemodels-activity-7106253824910139392-WZRM?utm_source=share&utm_medium=member_desktop) 623 | - [Overview of 🤗 Transformers Quantization: GPTQ vs bitsandbytes](https://huggingface.co/blog/overview-quantization-transformers) 624 | - [LoRA Exchange (LoRAX): Serve 100s of Fine-Tuned LLMs for the Cost of 1](https://predibase.com/blog/lora-exchange-lorax-serve-100s-of-fine-tuned-llms-for-the-cost-of-one) 625 | - [Introducing LoRAX](https://www.linkedin.com/posts/travisaddair_lora-exchange-lorax-serve-100s-of-fine-tuned-activity-7120819275442896896-vlI_?utm_source=share&utm_medium=member_desktop) 626 | - [DeepSparse: Sparsity-aware deep learning inference runtime for CPUs](https://github.com/neuralmagic/deepsparse) 627 | - [Practical Tips for Finetuning LLMs Using LoRA (Low-Rank Adaptation)](https://magazine.sebastianraschka.com/p/practical-tips-for-finetuning-llms) [**Great**] 628 | - [Dare method for improving LLMs performance](https://www.linkedin.com/posts/andrew-iain-jardine_llm-opensource-llms-activity-7134896163698208768-0Gyf?utm_source=share&utm_medium=member_desktop) 629 | - [Small model that surpass the GPT4](https://www.linkedin.com/posts/clementdelangue_open-models-now-starting-to-surpass-gpt4-activity-7137904570898264064-LSmc?utm_source=share&utm_medium=member_desktop) [Interesting] 630 | - [Efficient LLMs Survey](https://github.com/AIoT-MLSys-Lab/Efficient-LLMs-Survey) [Great] 631 | - [LoRAX (LoRA eXchange): Framework that allows users to serve thousands of fine-tuned models on a single GPU](https://github.com/predibase/lorax) 632 | - [PowerInfer: High-speed LLMs Serving on PCs with Consumer-grade GPUs](https://github.com/SJTU-IPADS/PowerInfer) 633 | - [LoRA From Scratch Implementation](https://www.linkedin.com/posts/sebastianraschka_code-lora-from-scratch-a-lightning-studio-activity-7155241298227060736-QRul?utm_source=share&utm_medium=member_desktop) 634 | - [Improving LoRA (DoRA): Implementing Weight-Decomposed Low-Rank Adaptation (DoRA)](https://www.linkedin.com/posts/sebastianraschka_improving-lora-implementing-weight-decomposed-activity-7165053172175024128-bqwu?utm_source=share&utm_medium=member_desktop) 635 | - [DoRA Link2](https://magazine.sebastianraschka.com/p/lora-and-dora-from-scratch) 636 | - [Proxy-Tuning (new method for fine-tuning LLMs)](https://www.linkedin.com/posts/sebastianraschka_theres-a-new-promising-method-for-finetuning-activity-7153788017017544705-ADC7?utm_source=share&utm_medium=member_desktop) 637 | - [AutoQuantize (GGUF, AWQ, EXL2, GPTQ) Colab Notebook](https://colab.research.google.com/drive/1Li3USnl3yoYctqJLtYux3LAIy4Bnnv3J?usp=sharing) [Great] 638 | - [DoRA: Weight-Decomposed Low-Rank Adaptation - Linkedin Post](https://www.linkedin.com/posts/sebastianraschka_while-everyone-is-talking-about-sora-theres-activity-7164268573756960770-N7Hu?utm_source=share&utm_medium=member_desktop) 639 | - [DoRA: Weight-Decomposed Low-Rank Adaptation - Paper](https://arxiv.org/abs/2402.09353) 640 | - [GaLore: Memory Efficient Fine-tuning Technique](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_galore-is-a-new-memory-efficient-fine-tuning-activity-7177599313294827521-kye2?utm_source=share&utm_medium=member_desktop) 641 | - [Quanto: a pytorch quantization toolkit](https://huggingface.co/blog/quanto-introduction) [**Great**] 642 | - [Quanto: Linkedin Post](https://www.linkedin.com/posts/dcorvoysier_quanto-a-pytorch-quantization-toolkit-activity-7175421050808078336-QcEM?utm_source=share&utm_medium=member_desktop) 643 | - [Deleting 40% of LLM Layers Without Drop in Accuracy](https://www.linkedin.com/posts/liorsinclair_researchers-just-developed-a-new-method-to-activity-7180929255411789826-z3TV?utm_source=share&utm_medium=member_desktop) 644 | - [The Unreasonable Ineffectiveness of the Deeper Layers](https://arxiv.org/html/2403.17887v1) 645 | - [Continual Pretraining of LLMs](https://www.linkedin.com/posts/sebastianraschka_we-talk-a-lot-about-finetuning-llms-to-follow-activity-7174395744068464642-jPFI?utm_source=share&utm_medium=member_desktop) 646 | - [NOLA: run 10,000 customized LLaMA2 (70B) (4bit) models on a single 48GB GPU](https://www.linkedin.com/posts/hpirsiav_iclr2024-iclr2024-activity-7192618595405725696-HZXu?utm_source=share&utm_medium=member_desktop) 647 | - [NOLA LLaMA3](https://www.linkedin.com/posts/s-hasan-abbas_syed-hasan-8503llama-3-8b-nola-hugging-activity-7193318944575762434-MD_T?utm_source=share&utm_medium=member_desktop) 648 | - [LoRA Learns Less and Forgets Less in comparision to full finetuning](https://www.linkedin.com/posts/sebastianraschka_lora-learns-less-and-forgets-less-when-i-activity-7197576220585201664-KA4L?utm_source=share&utm_medium=member_desktop) 649 | - [Best Practices for Fine-Tuning & Training LLMs](https://www.linkedin.com/posts/aleksagordic_amazing-list-of-techniques-for-improving-activity-7215624025639645184-496W?utm_source=share&utm_medium=member_android) 650 | - [TorchChat](https://www.linkedin.com/posts/pytorch_llms-mobilellms-localai-activity-7224090140011380737-RHdH?utm_source=share&utm_medium=member_desktop) 651 | - [The Evolution of Extreme LLM Compression: From QuIP to AQLM with PV-Tuning](https://medium.com/yandex/the-evolution-of-extreme-llm-compression-from-quip-to-aqlm-with-pv-tuning-19c44b91af96) 652 | - [Calculating GPU memory for serving LLMs](https://www.substratus.ai/blog/calculating-gpu-memory-for-llm) 653 | - [How Much GPU Memory is Needed to Serve a Large Language Model (LLM)?](https://masteringllm.medium.com/how-much-gpu-memory-is-needed-to-serve-a-large-languagemodel-llm-b1899bb2ab5d) 654 | - [CUDA-Free Inference for LLMs (PyTorch Blog)](https://pytorch.org/blog/cuda-free-inference-for-llms/?utm_content=306418724&utm_medium=social&utm_source=linkedin&hss_channel=lcp-78618366) 655 | - [The Ultra-Scale Playbook: Training LLMs on GPU Clusters](https://huggingface.co/spaces/nanotron/ultrascale-playbook) 656 | 657 | ### Productionizing LLMs: 658 | - [LLM From the Trenches: 10 Lessons Learned Operationalizing Models at GoDaddy](https://www.godaddy.com/resources/news/llm-from-the-trenches-10-lessons-learned-operationalizing-models-at-godaddy) 659 | 660 | ### LLMs on Mobile Devices: 661 | - [MLC LLM](https://github.com/mlc-ai/mlc-llm) 662 | 663 | ### LLM Applications & APIs: 664 | - [Building LLM applications for production](https://huyenchip.com/2023/04/11/llm-engineering.html) 665 | - [Bard API](https://github.com/dsdanielpark/Bard-API) 666 | - [Amazon Bedrock: build and scale generative AI applications](https://aws.amazon.com/bedrock/) [**Great**] 667 | 668 | ### Natural Language to SQL: 669 | - [text to SQL Github Repos](https://github.com/topics/text-to-sql) 670 | - [vanna](https://github.com/vanna-ai/vanna) 671 | - [sqlchat](https://github.com/sqlchat/sqlchat) 672 | - [dataherald](https://github.com/Dataherald/dataherald) 673 | - [WrenAI](https://github.com/Canner/WrenAI) 674 | - [Practical text-to-SQL for data analytics by Linkedin](https://www.linkedin.com/blog/engineering/ai/practical-text-to-sql-for-data-analytics) [Great] 675 | - [Persian abstract of above Practical text-to-SQL for data analytics by Linkedin - Out of Distribution Telegram Channel](https://t.me/out_of_distribution/1122) 676 | 677 | ### Prompt Engineering: 678 | - [Different Kinds of Prompt Engineering](https://www.linkedin.com/posts/munjal-patel_generativeai-largelanguagemodels-llm-activity-7051862874935197696-2E_J/?utm_source=share&utm_medium=member_android) 679 | - [Prompt Engineering Guide](https://www.promptingguide.ai/) 680 | - [PromptTools: tools for prompt testing and experimentation](https://github.com/hegelai/prompttools) 681 | - [Prompt engineering for Claude's long context window](https://www.anthropic.com/index/prompting-long-context) 682 | - [Chain of Verification Prompt engineering method](https://www.linkedin.com/posts/xamatriain_a-week-ago-meta-presented-a-new-prompt-engineering-activity-7114351307183820800-MsgT?utm_source=share&utm_medium=member_desktop) 683 | - [Analogical Prompting](https://huggingface.co/papers/2310.01714) 684 | - [Prompt Flow: Build high-quality LLM apps](https://github.com/microsoft/promptflow) 685 | - [Contrastive Chain-of-Thought Prompting (CCoT)](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_improve-chain-of-thought-prompting-by-adding-activity-7133477395944091648-TKlQ?utm_source=share&utm_medium=member_desktop) 686 | - [New Prompting Techniques](https://www.linkedin.com/posts/pramodith_promptengineering-llm-activity-7134507333530836992-evPU?utm_source=share&utm_medium=member_desktop) 687 | - [Openai Prompt Engineering Guide - Linkedin Post](https://www.linkedin.com/posts/eric-vyacheslav-156273169_game-changer-open-ai-just-released-their-activity-7141454141683343360-eunF?utm_source=share&utm_medium=member_desktop) 688 | - [Openai Prompt Engineering Guide](https://platform.openai.com/docs/guides/prompt-engineering) 689 | - [Anthropic Claude Metaprompt Tool](https://www.linkedin.com/posts/sahar-mor_anthropic-released-a-useful-tool-that-turns-activity-7194705248039444480-7KtG?utm_source=share&utm_medium=member_desktop) 690 | - [Anthropic Prompt Improver](https://www.anthropic.com/news/prompt-improver) 691 | - [Anthropic Prompt Improver Linkedin Post](https://www.linkedin.com/posts/anthropicresearch_weve-added-a-new-prompt-improver-to-the-activity-7262874194802036736-Q_RP?utm_source=share&utm_medium=member_desktop) 692 | - [Anthropic Evaluate Prompts Tool](https://www.anthropic.com/news/evaluate-prompts) 693 | - [Cohere Prompt Tuner: Prompt Optimization at Your Fingertips](https://cohere.com/blog/intro-prompt-tuner?utm_source=bensbites&utm_medium=newsletter&utm_campaign=daily-digest-talk-with-your-ai-besties) 694 | - [Quality Prompts: Use and evaluate prompting techniques quickly](https://github.com/sarthakrastogi/quality-prompts) 695 | - [Prompt Design at Character.AI](https://research.character.ai/prompt-design-at-character-ai/) 696 | - [Structured Prompting](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_structured-prompting-is-a-key-requirement-activity-7235928635633725440-0OG-?utm_source=share&utm_medium=member_desktop) 697 | - [Writing with AI: Five ways professional writers are leveraging ChatGPT](https://openai.com/chatgpt/use-cases/writing-with-ai/) 698 | - [Google Prompt Gallery](https://ai.google.dev/gemini-api/prompts) 699 | - [ell: The Language Model Programming Library](https://docs.ell.so/) 700 | - [Template prompts of Cursor & VS Code and etc](https://github.com/x1xhlol/system-prompts-and-models-of-ai-tools) [useful] 701 | - [System Prompts Leaks](https://github.com/asgeirtj/system_prompts_leaks/) 702 | 703 | ### LLM-based Recommender Systems: 704 | - [ChatGPT-based Recommender Systems](https://blog.reachsumit.com/posts/2023/05/chatgpt-for-recsys/) 705 | 706 | ### LLMs for Tabular Data: 707 | - [Unleashing the Potential of Large Language Models for Predictive Tabular Tasks in Data Science](https://arxiv.org/abs/2403.20208) 708 | - [LLMs for Tabular Data - Linkedin post](https://www.linkedin.com/posts/pascalbiese_unleashing-the-potential-of-llms-for-tabular-activity-7180873134743449600-ChWm?utm_source=share&utm_medium=member_desktop) 709 | 710 | ### LLMs as Classifiers (finetuning LLMs for classification): 711 | - [LLMs as Classifiers Linkedin Post1](https://www.linkedin.com/posts/sebastianraschka_what-if-you-care-about-finetuning-llms-for-activity-7183808393155944448-CSR1?utm_source=share&utm_medium=member_desktop) 712 | - [Training LLMs for Spam Classification](https://www.linkedin.com/posts/sebastianraschka_training-llms-for-spam-classification-i-activity-7197943692949676034-c6_j?utm_source=share&utm_medium=member_desktop) 713 | 714 | ### LLM Data Sets: 715 | - [SlimPajama: A 627B token cleaned and deduplicated version of RedPajama](https://www.cerebras.net/blog/slimpajama-a-627b-token-cleaned-and-deduplicated-version-of-redpajama) 716 | 717 | ### LLM based Agents: 718 | - [MetaGPT: Multi-Agent Framework](https://github.com/geekan/MetaGPT) 719 | - [DevOpsGPT: AI-Driven Software Development Automation Solution](https://github.com/kuafuai/DevOpsGPT) 720 | - [LLM Agent Survey](https://github.com/Paitesanshi/LLM-Agent-Survey) 721 | - [Microsoft AutoGen development of LLM applications using multiple agents](https://github.com/microsoft/autogen) 722 | - [OpenDevin: autonomous AI software engineer](https://github.com/OpenDevin/OpenDevin) 723 | - [Composio: the best toolset to integrate AI Agents](https://github.com/ComposioHQ/composio) 724 | - [MindSearch: An LLM-based Multi-agent Framework of Web Search Engine](https://github.com/InternLM/MindSearch) 725 | - [OpenAI Swarm Library for Multi-Agent](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_this-came-unexpected-openai-released-swarm-activity-7250841965519368192-oJ35?utm_source=share&utm_medium=member_desktop) 726 | - [Don't Sleep on Single-agent Systems](https://www.all-hands.dev/blog/dont-sleep-on-single-agent-systems) 727 | - [Linkedin post for Don't Sleep on Single-agent Systems](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_the-more-progress-we-make-on-llms-the-more-activity-7246758324912758784-VC3N?utm_source=share&utm_medium=member_desktop) 728 | - [Microsoft TinyTroupe library for simulate human agents with LLMs](https://www.linkedin.com/posts/sahar-mor_a-new-open-source-python-library-called-tinytroupe-activity-7262849272381874176-KFk_?utm_source=share&utm_medium=member_desktop) [Interesting] 729 | - [Google Whitepaper on AI Agents - Linkedin Post](https://www.linkedin.com/posts/eric-vyacheslav-156273169_whitepaper-ai-agents-ugcPost-7286059606814990338-JinO/?utm_source=share&utm_medium=member_desktop) 730 | - [Google Whitepaper on AI Agents](https://www.kaggle.com/whitepaper-agents) 731 | - [Microsoft ai-agents-for-beginners Course](https://github.com/microsoft/ai-agents-for-beginners) 732 | - [HuggingFace Smolagent Library blog post](https://huggingface.co/blog/smolagents) [Useful] 733 | 734 | ### Structured Output in LLMs: 735 | - [PydanticAI](https://github.com/pydantic/pydantic-ai) 736 | - [PydanticAI Linkedin Post](https://www.linkedin.com/posts/liorsinclair_theres-a-new-ai-agent-framework-that-lets-activity-7270122274408534017-OOQq?utm_source=share&utm_medium=member_desktop) 737 | 738 | ### Deploying LLMs: 739 | - [ExecuTorch Post1](https://www.linkedin.com/posts/pytorch_introducing-executorch-alpha-executorch-activity-7191120577749831680-vYzE?utm_source=share&utm_medium=member_desktop) 740 | 741 | ### LLM Engineering: 742 | - [Langfuse: Open Source LLM Engineering Platform](https://github.com/langfuse/langfuse) 743 | 744 | ### External Tools that Useful for LLMs: 745 | - [Microsoft MarkItDown: Python library that lets you convert any document to Markdown](https://www.linkedin.com/posts/liorsinclair_microsoft-just-open-sourced-markitdown-a-activity-7275201481828454403-c5TX?utm_source=share&utm_medium=member_desktop) [Great] 746 | 747 | ### Notes about Cost & Price of Training and Using LLMs: 748 | - [Cost to Deploy LLaMA2 vs. ChatGPT](https://www.linkedin.com/posts/damienbenveniste_machinelearning-datascience-artificialintelligence-activity-7109561666324885504-ySeC?utm_source=share&utm_medium=member_desktop) [Very Important] 749 | - [Anyscale Training Cost](https://www.linkedin.com/posts/robert-nishihara-b6465444_im-so-proud-of-what-we-launched-last-week-activity-7113021412084219904-WFbP?utm_source=share&utm_medium=member_desktop) 750 | - [LLMs APIs Pricing Benchmark: pricing of AWS Bedrock, OpenAI, Microsoft Azure](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_yesterday-amazon-web-services-aws-released-activity-7113454144216031233-LYuF?utm_source=share&utm_medium=member_desktop) 751 | - [LLM Token-based Price Sheet](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_claude-21-with-200k-context-just-got-released-activity-7132812689369657344-Rk_a?utm_source=share&utm_medium=member_desktop) 752 | - [LLM Pricing Table Sheet](https://docs.google.com/spreadsheets/d/1NX8ZW9Jnfpy88PC2d6Bwla87JRiv3GTeqwXoB4mKU_s/edit#gid=0) 753 | - [LLM Pricing Table Linkedin Post](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_updated-llm-pricing-table-earlier-today-activity-7170527176168042497-YgT4?utm_source=share&utm_medium=member_desktop) 754 | - [Pricibg Sheet for Hosted LLMs](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_just-updated-my-pricing-sheet-for-hosted-activity-7213556290575368196-u71R?utm_source=share&utm_medium=member_desktop) 755 | - [LLM Pricing Comparison Tool in HuggingFace Space](https://huggingface.co/spaces/philschmid/llm-pricing) 756 | 757 | ### Excellent & Easy to Learn Resources for Learning Transformers: 758 | - [e2eml transformers from scratch](https://e2eml.school/transformers.html) [**Excellent**] 759 | - [annotated-transformer: Learning transformers from code](http://nlp.seas.harvard.edu/annotated-transformer/#a-first-example) 760 | - [Transformers Recipe](https://github.com/dair-ai/Transformers-Recipe) 761 | 762 | ### Persian based Transformer Models: 763 | - [ALBERT-Persian](https://github.com/m3hrdadfi/albert-persian) 764 | - [ALBERT-Persian Demo Page](https://albert-lab.m3hrdadfi.me/) 765 | - [ALBERT-Farsi-base-v2 in HuggingFace](https://huggingface.co/m3hrdadfi/albert-fa-base-v2) 766 | - [ParsBERT - Model for Persian Language Understanding](https://github.com/hooshvare/parsbert) 767 | - [ARMAN](https://github.com/alirezasalemi7/ARMAN) [Great] 768 | - [ParsBigBird: Persian Bert For Long-Range Sequences](https://github.com/sajjjadayobi/ParsBigBird) [Great] 769 | - [PersianQA](https://github.com/sajjjadayobi/PersianQA) 770 | - [Persian (Farsi) Pre-trained Language Models](https://nlpdataset.ir/farsi/pre-trained_lm.html) [Great] 771 | - [Hezar: The all-in-one AI library for Persian, supporting a wide variety of tasks and modalities](https://github.com/hezarai/hezar) [**Great & Important**] 772 | - [XLM-RoBERTa (Multilingual & supports Persian)](https://huggingface.co/FacebookAI/xlm-roberta-base) 773 | - [TookaBERT by PartAI](https://huggingface.co/PartAI/TookaBERT-Large) [Great] 774 | - [Dorna PartAI LLM](https://www.linkedin.com/posts/partdp-ai_aetaexaesabraeaaeqaepaeuahy-aevaewaecaetaedaeuaewaehahy-activity-7205158585968844800-sqqa/?utm_source=share&utm_medium=member_desktop) 775 | 776 | ## Transfer Learning with Transformers: 777 | - [Transfer Learning for NLP via BERT for Text Classification](https://www.analyticsvidhya.com/blog/2020/07/transfer-learning-for-nlp-fine-tuning-bert-for-text-classification/) 778 | - [Text Classification with BERT Tokenizer](https://stackabuse.com/text-classification-with-bert-tokenizer-and-tf-2-0-in-python/) 779 | - [Bert Text Classification](https://github.com/Shivampanwar/Bert-text-classification) 780 | - [Persian Semantic Search](https://github.com/m3hrdadfi/semantic-search) 781 | - [Toward fine-tuning a state of the art Natural Language Inference (NLI) model for Persian](https://haddadhesam.medium.com/toward-fine-tuning-a-state-of-the-art-natural-language-inference-nli-model-for-persian-4d538ea4525d) 782 | 783 | ### Siamese Netowrks and Dual BERT for Multi Text Classification: 784 | - [Siamese and Dual BERT for Multi-text Classification](https://towardsdatascience.com/siamese-and-dual-bert-for-multi-text-classification-c6552d435533) 785 | - [Transfer Learning via Siamese Networks](https://www.inovex.de/blog/transfer-learning-siamese-networks/) 786 | 787 | ## Attention Mechanism: 788 | - [Attention Mechanism](https://blog.floydhub.com/attention-mechanism/) 789 | - [Visualizing A Neural Machine Translation Model - Attention Mechanism](https://jalammar.github.io/visualizing-neural-machine-translation-mechanics-of-seq2seq-models-with-attention/) 790 | - [Intuitive Understanding of Attention Mechanism in Deep Learning](https://towardsdatascience.com/intuitive-understanding-of-attention-mechanism-in-deep-learning-6c9482aecf4f) 791 | - [Structured Attention Networks](https://medium.com/uci-nlp/summary-structured-attention-networks-f1917dd622af) 792 | 793 | ## Sequence Modeling: 794 | - [WaveNet: Increasing reception field using dilated convolution](https://medium.com/@kion.kim/wavenet-a-network-good-to-know-7caaae735435) 795 | - [Understanding WaveNet architecture](https://medium.com/@satyam.kumar.iiitv/understanding-wavenet-architecture-361cc4c2d623) 796 | - [WaveNet: A Generative Model for Raw Audio](https://medium.com/a-paper-a-day-will-have-you-screaming-hurray/wavenet-a-generative-model-for-raw-audio-84b2aa5fb4a0) 797 | - [How WaveNet Works](https://towardsdatascience.com/how-wavenet-works-12e2420ef386) 798 | - [PyTorch Tutorial to Sequence Labeling](https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Sequence-Labeling) 799 | 800 | ## Text Summarization: 801 | - [Bert Extractive Summarizer](https://pypi.org/project/bert-extractive-summarizer/) [**Great**] 802 | - [Generating Text Summaries Using GPT-2 on PyTorch with Minimal Training](https://blog.paperspace.com/generating-text-summaries-gpt-2/) [_Good_] 803 | - [A Gentle Introduction to Text Summarization in Machine Learning](https://blog.floydhub.com/gentle-introduction-to-text-summarization-in-machine-learning/) 804 | - [Taming Recurrent Neural Networks for Better Summarization](http://www.abigailsee.com/2017/04/16/taming-rnns-for-better-summarization.html) 805 | - [PyTorch implementation of "Get to the point"](https://github.com/mjc92/GetToThePoint) 806 | - [TensorFlow implementation of "Get to the point"](https://github.com/abisee/pointer-generator) 807 | 808 | ## Language Model: 809 | - [A Comprehensive Guide to Build your own Language Model in Python](https://www.analyticsvidhya.com/blog/2019/08/comprehensive-guide-language-model-nlp-python-code/) 810 | - [D2L: Language Models and Dataset](https://d2l.ai/chapter_recurrent-neural-networks/language-models-and-dataset.html) 811 | - [Develop a word-level Neural Language Model in Keras](https://machinelearningmastery.com/how-to-develop-a-word-level-neural-language-model-in-keras/) 812 | - [IBM deep learning language model](https://github.com/IBM/deep-learning-language-model) 813 | - [BERT language model](https://devopedia.org/bert-language-model) 814 | - [Facebook AI: GSLM](https://www.marktechpost.com/2021/09/09/facebook-ai-introduces-gslm-generative-spoken-language-model-a-textless-nlp-model-that-breaks-free-completely-of-the-dependence-on-text-for-training/) 815 | - [Language Modeling Great Tutorial](https://lena-voita.github.io/nlp_course/language_modeling.html) 816 | - [GALACTICA: general-purpose scientific language model](https://github.com/paperswithcode/galai) [Great] 817 | - [Distributed Training of Language Models with Reinforcement Learning via Human Feedback (RLHF)](https://github.com/CarperAI/trlx) [**Excellent**] 818 | 819 | ## Text & Document Classification: 820 | - [hedwig - PyTorch deep learning models for document classification](https://github.com/castorini/hedwig) 821 | 822 | ## Topic Modeling: 823 | - [Topic Modeling with BERT](https://towardsdatascience.com/topic-modeling-with-bert-779f7db187e6) 824 | - [BERTopic: Great Library for Topic Modeling](https://github.com/MaartenGr/BERTopic) [Great] 825 | 826 | ## Sentiment Analysis: 827 | - [Introduction to Deep Learning – Sentiment Analysis](https://nlpforhackers.io/deep-learning-introduction/) 828 | 829 | ## Co-Reference Resolution: 830 | - [Coreference Resolution for Chatbots](https://medium.com/huggingface/state-of-the-art-neural-coreference-resolution-for-chatbots-3302365dcf30) 831 | - [Hugging Face - CoRef](https://huggingface.co/coref/) 832 | 833 | ## Imbalance Handling in NLP: 834 | - [Over-Sampling using SMOTE](https://imbalanced-learn.readthedocs.io/en/stable/generated/imblearn.over_sampling.SMOTE.html) [_SMOTE for high-dimensional class-imbalanced data_] 835 | - [Over-sampling via imbalanced-learn library](https://imbalanced-learn.readthedocs.io/en/stable/over_sampling.html) 836 | - [Imbalanced Data Handling](https://www.jeremyjordan.me/imbalanced-data/) 837 | 838 | ## Information Retrieval: 839 | - [PyTerrier: Python API for Terrier](https://github.com/terrier-org/pyterrier) 840 | 841 | ## Distance Measures: 842 | - [Edit Distance](https://www.geeksforgeeks.org/edit-distance-dp-5/) 843 | 844 | ## Text-based Emotion Recognition: 845 | - [XLM-EMO: Multilingual Emotion Prediction in Social Media Text](https://github.com/MilaNLProc/xlm-emo) 846 | 847 | ## Machine Translation: 848 | - [Open-NLLB: No Language Left Behind (NLLB), models capable of delivering high-quality translations directly between any pair of 200+ languages](https://github.com/gordicaleksa/Open-NLLB) 849 | 850 | ## Chatbot: 851 | - [Rasa Chatbot](https://github.com/RasaHQ/rasa) [**Great**] 852 | - [Learn how to Build and Deploy a Chatbot in Minutes using Rasa](https://www.analyticsvidhya.com/blog/2019/04/learn-build-chatbot-rasa-nlp-ipl/) 853 | - [chatbot with DialoGPT](https://www.machinecurve.com/index.php/2021/03/16/easy-chatbot-with-dialogpt-machine-learning-and-huggingface-transformers/) 854 | - [DialoGPT: huggingface Transformer](https://huggingface.co/transformers/model_doc/dialogpt.html) 855 | - [deeppavlov](https://github.com/deeppavlov/DeepPavlov) [**Great**] 856 | - [PyTorch Chatbot Tutorial](https://brsoff.github.io/tutorials/beginner/chatbot_tutorial.html) 857 | - [Implement a Simple Chat Bot With PyTorch](https://www.python-engineer.com/posts/chatbot-pytorch/) 858 | - [GPT2 Chatbot PyTorch](https://github.com/devjwsong/gpt2-chatbot-pytorch) 859 | - [PyTorch Official Chatbot Tutorial](https://pytorch.org/tutorials/beginner/chatbot_tutorial.html) 860 | - [PaddlePaddle Knover: toolkit for knowledge grounded dialogue generation](https://github.com/PaddlePaddle/Knover) 861 | - [PaddlePaddle PLATO-2](https://github.com/PaddlePaddle/PaddleNLP/tree/develop/examples/dialogue/plato-2) 862 | - [ParlAI](https://github.com/facebookresearch/ParlAI) [Great] 863 | - [huggingface: Transformers](https://github.com/huggingface/transformers) [Great] 864 | - [huggingface: Blenderbot](https://huggingface.co/transformers/model_doc/blenderbot.html) [**Great**] 865 | - [huggingface: Blenderbot Small](https://huggingface.co/transformers/model_doc/blenderbot_small.html) [**Great**] 866 | - [huggingface: GPT-2 Text Generation](https://huggingface.co/gpt2?text=A+long+time+ago%2C) [**Great**] 867 | - [Seq2seq Chatbot](https://github.com/ricsinaruto/Seq2seqChatbots) 868 | - [seq2seq Chatbot implemented in Pytorch](https://github.com/khordoo/chatbot-pytorch) 869 | - [papers with code: chatbot](https://paperswithcode.com/task/chatbot) 870 | - [Proudly Leading the Chatbot](https://www.analyticsinsight.net/ankush-sabharwal-proudly-leading-the-chatbot-sphere-with-strategical-innovations-and-implementations/) 871 | - [Real Python: Build a Chatbot with Python ChatterBot](https://realpython.com/build-a-chatbot-python-chatterbot/) 872 | - [A step-by-step guide to building a chatbot based on your own documents with GPT](https://bootcamp.uxdesign.cc/a-step-by-step-guide-to-building-a-chatbot-based-on-your-own-documents-with-gpt-2d550534eea5) 873 | - [MiniPerplx: an alternative to Perplexity that lets search the web, research papers, youtube videos, movies](https://scira.app/) 874 | - [GitHub Models](https://github.blog/news-insights/product-news/introducing-github-models/) 875 | - [Git Ingest: Quickly turn a GitHub repository into text for LLMs](https://www.linkedin.com/posts/eric-vyacheslav-156273169_you-can-now-quickly-turn-a-github-repository-activity-7277322180223254528-CRW9?utm_source=share&utm_medium=member_desktop) [**Great**] 876 | - [Create a Chatbot for any GitHub repo](https://www.linkedin.com/posts/eric-vyacheslav-156273169_game-changer-you-can-now-create-a-chatbot-activity-7226604741261230081-Bthf?utm_source=share&utm_medium=member_desktop) [**Great**] 877 | 878 | ### Chatbot & LLMs Evaluation Metrics: 879 | - [Chatbot Analytics: 9 Key Metrics](https://www.tidio.com/blog/chatbot-analytics/) 880 | - [Chatbot Statistics for 2023](https://www.tidio.com/blog/chatbot-statistics/) 881 | - [Chatbot Analytics 101: Essential Metrics to Track](https://blog.hootsuite.com/chatbot-analytics/) 882 | - [12 Metrics For Chatbot Analytics](https://www.kommunicate.io/blog/metrics-for-chatbot-analytics/) 883 | - [ParlAI Evaluation Metrics for Chatbot](https://github.com/facebookresearch/ParlAI/blob/14a10258bf90218341e0253d1c5a88c9d2cd013f/docs/source/tutorial_metrics.md) 884 | - [Chatbot Evaluation Metrics](https://github.com/ahkarami/Great-Deep-Learning-Tutorials/blob/master/NLP/Chatbot_Evaluation_Metrics.md) [**Great**] 885 | - [Databricks' report on LLM evaluation methods](https://www.linkedin.com/posts/activity-7107825117379907584-m17h?utm_source=share&utm_medium=member_desktop) 886 | - [AgentBench: Evaluating LLMs as Agents](https://github.com/THUDM/AgentBench) 887 | - [Prometheus: Using GPT4 as SLMs Evaluator](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_using-powerful-llms-gpt-4-as-an-evaluator-activity-7131951255119110145-RH86?utm_source=share&utm_medium=member_desktop) 888 | - [LLM Model Evaluation Metrics - When and How to Use Them](https://www.linkedin.com/posts/amrita-rath-288a071bb_llm-evaluation-metrics-activity-7198262398464503808-Gs6y?utm_source=share&utm_medium=member_desktop) 889 | 890 | ### OpenAI ChatGPT & Its Applications: 891 | - [OpenAI ChatGPT](https://openai.com/blog/chatgpt/) [Amazing] 892 | - [Description of How OpenAI ChatGPT Works: Illustrating Reinforcement Learning from Human Feedback (RLHF)](https://github.com/huggingface/blog/blob/main/rlhf.md) 893 | - [How ChatGPT was Trained](https://www.linkedin.com/posts/damienbenveniste_machinelearning-datascience-chatgpt-activity-7007019154666909696-T5WM/?utm_source=share&utm_medium=member_android) 894 | - [ChatGPT Android SDK](https://github.com/skydoves/chatgpt-android/releases) 895 | - [ChatGPT awesome apps](https://www.linkedin.com/posts/tarrysingh_chatgpt-activity-7017947289721655296-7-pK/?utm_source=share&utm_medium=member_android) 896 | - [A Categorical Archive of ChatGPT Failures](https://arxiv.org/abs/2302.03494) 897 | - [Is ChatGPT a General-Purpose Natural Language Processing Task Solver?](https://arxiv.org/abs/2302.06476) 898 | - [aman.ai chatGPT Tutorial](https://aman.ai/primers/ai/chatGPT/) [Great] 899 | - [ChatGPT for customer service](https://www.intercom.com/ai-bot) 900 | - [Chatgpt Retrieval Plugin](https://github.com/openai/chatgpt-retrieval-plugin) 901 | - [Trending AI Tools](https://galionaitools.blogspot.com/2023/03/trending-ai-tools.html) 902 | - [Merlin: OpenAI ChatGPT Plus extension on all websites](https://merlin.foyer.work/) 903 | - [Adrenaline](https://useadrenaline.com/app) 904 | - [Using LLMs as agents that orchestrate tools](https://www.linkedin.com/posts/moritz-laurer_augmented-language-models-a-survey-activity-7047924951625953281-0XDj/?utm_source=share&utm_medium=member_android) [Interesting] 905 | - [ChatGPT API Using Python](https://www.machinelearning-basics.com/2023/04/chatgpt-api-using-python.html?m=1) 906 | - [parthean: A Startup about Financial Expert via ChatGPT](https://www.parthean.com/) 907 | - [Notes on the cost of ChatGPT](https://www.linkedin.com/posts/laurencevanelegem_sam-altman-ceo-of-openai-dropped-a-at-activity-7061987804548870144-RF9y/?utm_source=share&utm_medium=member_android) 908 | - [Ortus - your YouTube AI buddy](https://chrome.google.com/webstore/detail/ortus-your-youtube-ai-bud/jmpepfdhkjkknfpnfohnmnjoceepcbmp) 909 | - [How Is ChatGPT’s Behavior Changing over Time?](https://www.linkedin.com/posts/svpino_gpt-4-is-getting-worse-over-time-not-better-activity-7087379892077481984-uORp?utm_source=share&utm_medium=member_android) 910 | - [LLM Drifts: How Is ChatGPT’s Behavior Changing over Time?](https://github.com/lchen001/LLMDrift) 911 | - [ChatGPT app Builder](https://www.linkedin.com/posts/zainkahn_absolute-madness-openai-ceo-sam-altman-activity-7128011745868050432-Ox5K?utm_source=share&utm_medium=member_desktop) 912 | - [GPT4 Turbo 128k analysis Notes (its price)](https://www.linkedin.com/posts/reuvencohen_i-finally-got-a-chance-to-play-with-the-new-activity-7128179916512104448-SlEX?utm_source=share&utm_medium=member_desktop) 913 | - [Designer GPT: website creator](https://www.linkedin.com/posts/eric-vyacheslav-156273169_this-is-crazy-designergpt-is-a-new-gpt-that-activity-7129833701873438720-lQuN?utm_source=share&utm_medium=member_desktop) 914 | - [OpenAI DevDay Breakout Sessions Videos](https://www.linkedin.com/posts/openai_openai-devday-breakout-sessions-youtube-activity-7130298061599195137-vbyY?utm_source=share&utm_medium=member_desktop) 915 | - [GPT Seed Parameter Notes](https://www.linkedin.com/posts/sahar-mor_openai-released-a-feature-that-mitigates-activity-7130940108974788608-vkDW?utm_source=share&utm_medium=member_desktop) 916 | - [Awesome ChatGPT Prompts](https://github.com/f/awesome-chatgpt-prompts) 917 | - [GPT-4o Full Data Analysis](https://www.linkedin.com/posts/eric-vyacheslav-156273169_gpt-4o-can-do-full-data-analysis-from-a-single-activity-7196162441116860416--yzu?utm_source=share&utm_medium=member_desktop) 918 | - [GPT4-o Architecture](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_is-this-the-architecture-of-openai-gpt-4o-activity-7199759664836739073-gTEz?utm_source=share&utm_medium=member_desktop) 919 | - [Introducing Structured Outputs in the OpenAI API](https://openai.com/index/introducing-structured-outputs-in-the-api/) 920 | - [OpenAI Realtime-api](https://openai.com/index/introducing-the-realtime-api/) 921 | - [OpenAI Model Distillation in the API](https://openai.com/index/api-model-distillation/) 922 | - [OpenAI Prompt Caching](https://platform.openai.com/docs/guides/prompt-caching) 923 | - [LibreChat: Enhanced ChatGPT Clone](https://github.com/danny-avila/LibreChat) [**Great**] 924 | 925 | ### OpenAI Learning to Reason & O1 Models: 926 | - [Learning to Reason with LLMs: OpenAI o1 Model](https://openai.com/index/learning-to-reason-with-llms/) 927 | - [How does OpenAI train the Strawberry (o1) model to spend more time thinking?](https://www.linkedin.com/posts/tom-yeh_openai-strawberry-aibyhand-activity-7240201012697833472-rrzD?utm_source=share&utm_medium=member_desktop) 928 | - [Learning to Reason before you speak is how OpenAI o1 generates its response](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_learning-to-reason-before-you-speak-is-how-activity-7240629908559785984--wMj?utm_source=share&utm_medium=member_desktop) 929 | - [5 Papers that better understanding Openai o1 models](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_here-are-5-papers-you-want-to-read-to-understand-activity-7241017716214571008-eVba/?utm_source=share&utm_medium=member_android) 930 | 931 | ## Google Bard & Gemini: 932 | - [Google DeepMind Gemini](https://www.linkedin.com/posts/googledeepmind_introducing-gemini-googles-largest-and-activity-7138182085441118208--M-h?utm_source=share&utm_medium=member_desktop) 933 | - [Google released Gemini](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_google-just-released-gemini-their-most-activity-7138191392861757440-djDD?utm_source=share&utm_medium=member_desktop) 934 | - [Google Gemini official released notes](https://blog.google/technology/ai/google-gemini-ai/?utm_source=linkedin&utm_medium=social&utm_campaign=GDMGemini) 935 | 936 | ## Anthropic Claude: 937 | - [Anthropic Claude Tool Use](https://www.linkedin.com/posts/anthropicresearch_tool-use-is-now-available-in-beta-to-all-activity-7201976267171086336-oQ4K?utm_source=share&utm_medium=member_desktop) 938 | - [Anthropic Prompt Generator](https://www.linkedin.com/posts/liorsinclair_anthropic-mightve-just-solved-prompt-engineering-activity-7196911121939795968-yray?utm_source=share&utm_medium=member_desktop) 939 | - [Switched to Claude 3.5](https://www.interconnects.ai/p/switched-to-claude-from-chatgpt) 940 | - [Anthropic Message Batches API](https://www.anthropic.com/news/message-batches-api) 941 | - [Anthropic Message Batches API - Linkdin Post](https://www.linkedin.com/posts/anthropicresearch_introducing-the-message-batches-api-activity-7249461524996440066-xS37?utm_source=share&utm_medium=member_desktop) 942 | - [OpenAI Prompt Caching in GPT 4o and o1: How Does It Compare To Claude Prompt Caching?](https://blog.getbind.co/2024/10/03/openai-prompt-caching-how-does-it-compare-to-claude-prompt-caching/) 943 | - [Anthropic Blog: Transformer Circuits Thread](https://transformer-circuits.pub/) 944 | - [Anthropic MCP (Model Context Protocol)](https://modelcontextprotocol.io/quickstart) 945 | 946 | ## How do LLMs think? 947 | - [On the Biology of a Large Language Model](https://transformer-circuits.pub/2025/attribution-graphs/biology.html) 948 | 949 | ## NLP Programming Notes: 950 | - [100 Times Faster Natural Language Processing in Python](https://medium.com/huggingface/100-times-faster-natural-language-processing-in-python-ee32033bdced) 951 | - [Multi-label Text Classification using BERT](https://medium.com/huggingface/multi-label-text-classification-using-bert-the-mighty-transformer-69714fa3fb3d) 952 | - [Learning Meaning in Natural Language Processing](https://medium.com/huggingface/learning-meaning-in-natural-language-processing-the-semantics-mega-thread-9c0332dfe28e) 953 | - [Train and Deploy the Mighty Transformer NLP models using FastBert and AWS SageMaker](https://medium.com/@kaushaltrivedi/train-and-deploy-mighty-transformer-nlp-models-using-fastbert-and-aws-sagemaker-cc4303c51cf3) 954 | - [Distilling knowledge from Neural Networks to build smaller and faster models](https://blog.floydhub.com/knowledge-distillation/) 955 | - [HarfBuzz - a text shaping library](https://github.com/harfbuzz/harfbuzz) [_Useful_] 956 | - [PruneBERT - Hugging Face](https://github.com/huggingface/transformers/tree/master/examples/movement-pruning) 957 | - [spacy-streamlit: spaCy building blocks for Streamlit apps](https://github.com/explosion/spacy-streamlit) 958 | - [HuggingFace Evaluate Library](https://github.com/huggingface/evaluate) 959 | - [NeMo - toolkit for Conversational AI](https://github.com/NVIDIA/NeMo) [_Excellent_] 960 | 961 | ## Data Annotation Tools: 962 | - [doccano is an open source text annotation tool](https://github.com/doccano/doccano) [**Great**] 963 | - [doccano-divar](https://doccano.divar.ir/) 964 | 965 | ## Dataset Creator Tools: 966 | - [Nvidia create dataset from massive pdf files tool](https://www.linkedin.com/posts/liorsinclair_nvidia-just-released-a-powerful-pdf-extraction-ugcPost-7267580522359336962-GAQv?utm_source=share&utm_medium=member_android) 967 | 968 | ## NLP Courses: 969 | - [HuggingFace Course](https://github.com/huggingface/course) 970 | - [NLP Zero to One: Full Course](https://medium.com/nerd-for-tech/nlp-zero-to-one-full-course-4f8e1902c379) 971 | - [Stanford CS25: Transformers United](https://web.stanford.edu/class/cs25/) 972 | 973 | ## Other NLP Topics & miscellaneous: 974 | - [HybridNLP - Tutorial on Hybrid Techniques for Knowledge-based NLP](https://github.com/hybridnlp/tutorial) 975 | - [Top 10 GPT-3 Tools Easing Content Creation Work in 2022](https://www.analyticsinsight.net/top-10-gpt-3-tools-easing-content-creation-work-in-2022/) [Interesting] 976 | - [Inflection-2.5 CahtBot](https://inflection.ai/inflection-2-5) 977 | - [Research Paper Report Generating Agent](https://github.com/run-llama/llamacloud-demo/blob/main/examples/report_generation/research_paper_report_generation.ipynb) 978 | - [Fast Semantic Text Deduplication](https://www.linkedin.com/posts/patrick-fleith_2-lines-of-code-to-deduplicate-a-dataset-activity-7289903818069164032-p2aa?utm_source=share&utm_medium=member_android) 979 | -------------------------------------------------------------------------------- /NLP/Chatbot_Evaluation_Metrics.md: -------------------------------------------------------------------------------- 1 | ## Context 2 | Chatbot evaluation metrics can be examined from three different perspectives: 1) metrics related to AI-based chatbots, 2) metrics related to generative AI based chatbots, and 3) metrics to evaluate a complete chatbot system. In the following, I will introduce the criteria of each. 3 | 4 | ## Common evaluation metrics for AI-based chatbots 5 | 6 | The common evaluation metrics for AI-based chatbots include: 7 | 8 | - **Perplexity:** a measure of how well a language model predicts the next word in a sentence. It is calculated as the exponentiated average log-probability of the test set, with lower values indicating a better model [1, 2]. 9 | 10 | - **BLEU:** BLEU (Bilingual Evaluation Understudy) is a measure of the similarity between the predicted response and the reference response. It compares the n-grams (i.e. sequence of words) in the predicted response to the n-grams in the reference response, and calculates a score based on how many n-grams match. A higher BLEU score indicates a higher degree of similarity between the predicted response and the reference response [3-5]. 11 | 12 | - **METEOR:** METEOR (Metric for Evaluation of Translation with Explicit ORdering) is a measure of the similarity between the predicted response and the reference response, which also takes into account synonyms and stemming. It uses an alignment algorithm to align the words in the predicted response and reference response, and calculates a score based on how many aligned words match. A higher METEOR score indicates a higher degree of similarity between the predicted response and the reference response [6-8]. 13 | 14 | - **ROUGE:** ROUGE (Recall-Oriented Understudy for Gisting Evaluation) is a measure of the similarity between the predicted response and the reference response, which can be used for summarization tasks. It compares the n-grams in the predicted response to the n-grams in the reference response, and calculates a recall score based on how many n-grams in the predicted response also appear in the reference response. A higher ROUGE score indicates a higher degree of similarity between the predicted response and the reference response [9-11]. 15 | 16 | - **Embedding Average cosine Similarity (EACS):** EACS (Embedding Average cosine Similarity) is a measure of the semantic similarity between the predicted response and the reference response, which is based on word embeddings. It calculates the cosine similarity between the average word embeddings of the predicted response and the reference response. A higher EACS score indicates a higher degree of semantic similarity between the predicted response and the reference response [12-16]. 17 | 18 | - **Intent accuracy:** Intent accuracy is a measure of how well the chatbot recognizes the user's intent. It is calculated as the proportion of user inputs that are correctly classified by the chatbot's intent recognition module. 19 | 20 | - **Task completion rate:** Task completion rate is a measure of how well the chatbot is able to complete the task assigned to it. It is calculated as the proportion of user inputs for which the chatbot is able to provide a correct and complete response. 21 | 22 | 23 | ## Common evaluation metrics for Generative AI-based chatbots 24 | 25 | For Generative AI-based chatbots, such as Blenderbot or DialoGPT, some additional evaluation metrics that are commonly used include: 26 | 27 | - **Distinct n-gram:** Distinct n-gram is a measure of the diversity of the generated text. It is calculated as the proportion of unique n-grams (i.e. sequences of words) in the generated text. The higher the distinct-n value, the more diverse the generated text is, and hence more likely to be human-like. 28 | 29 | - **Self-BLEU:** Self-BLEU is a measure of the similarity of the generated text to the text it was trained on. It compares the n-grams in the generated text to the n-grams in the training text, and calculates a score based on how many n-grams match. A higher Self-BLEU score indicates a higher degree of similarity between the generated text and the training text. 30 | 31 | - **Fraternity:** Fraternity is a measure of the similarity of the generated text to the text generated by other models trained on the same data. It compares the generated text to the text generated by other models, and calculates a score based on how many n-grams match. A higher Fraternity score indicates a higher degree of similarity between the generated text and the text generated by other models. 32 | 33 | - **Human evaluation:** Human evaluation is a measure of the quality of the generated text as judged by human evaluators. This can include metrics such as coherence, fluency, and overall quality. The human evaluation can be done through surveys, where human evaluators rate the quality of the generated text on a scale or through the Turing test where the human evaluators are asked to identify if the text was generated by a machine or human. 34 | 35 | - **Adversarial evaluation:** Adversarial evaluation is a measure of the model's ability to generate text that is indistinguishable from text written by humans. This is achieved by training a discriminator model to distinguish between human-written text and machine-generated text, and then using the discriminator to evaluate the quality of the generated text. A high performance on adversarial evaluation indicates that the generated text is difficult to distinguish from human-written text. 36 | 37 | 38 | ## Common evaluation metrics for a whole chatbot system 39 | 40 | When evaluating a whole chatbot system, which includes both the natural language understanding (NLU) and natural language generation (NLG) components, common evaluation metrics include: 41 | 42 | - **Task completion rate:** Task completion rate is a measure of how well the chatbot is able to complete the task assigned to it. It is calculated as the proportion of user inputs for which the chatbot is able to provide a correct and complete response. 43 | 44 | - **Intent accuracy:** Intent accuracy is a measure of how well the chatbot recognizes the user's intent. It is calculated as the proportion of user inputs that are correctly classified by the chatbot's intent recognition module. 45 | 46 | - **Dialogue-level metrics:** a measure of how well the chatbot is able to carry out a dialogue with the user, such as dialogue length (i.e., the number of turns in a dialogue), and the number of successful dialogues (i.e., the number of dialogues that successfully reach the desired outcome). 47 | 48 | - **User satisfaction:** a measure of how satisfied the users are with the chatbot, which can be assessed through surveys or interviews. It can include metrics such as: Chatbot usability, Chatbot helpfulness, and Chatbot friendliness. 49 | 50 | - **Engagement:** a measure of how long the users spend interacting with the chatbot, how many interactions they have, and how much they return to the chatbot. It can include metrics such as: Session duration, Number of interactions, and Return rate. 51 | 52 | - **Retention:** a measure of how many users return to the chatbot after their first interaction. It can be calculated as the proportion of users who return to the chatbot after their first interaction. 53 | 54 | - **Business-level metrics:** a measure of how well the chatbot is able to meet the business objectives, such as customer service efficiency (i.e., measure of how quickly the chatbot is able to respond to customer queries), conversion rate (i.e., measure of how many users complete a desired action (e.g., making a purchase)), and revenue (measure of how much revenue is generated by the chatbot). 55 | 56 | ## References: 57 | - [1](https://en.wikipedia.org/wiki/Perplexity) 58 | - [2](https://medium.com/nlplanet/two-minutes-nlp-perplexity-explained-with-simple-probabilities-6cdc46884584) 59 | - [3](https://cloud.google.com/translate/automl/docs/evaluate#:~:text=BLEU%20(BiLingual%20Evaluation%20Understudy)%20is,of%20high%20quality%20reference%20translations.) 60 | - [4](https://en.wikipedia.org/wiki/BLEU) 61 | - [5](https://www.geeksforgeeks.org/nlp-bleu-score-for-evaluating-neural-machine-translation-python/) 62 | - [6](https://huggingface.co/spaces/evaluate-metric/meteor) 63 | - [7](https://en.wikipedia.org/wiki/METEOR) 64 | - [8](https://machinelearninginterview.com/topics/machine-learning/meteor-for-machine-translation/) 65 | - [9](https://en.wikipedia.org/wiki/ROUGE_(metric)) 66 | - [10](https://huggingface.co/spaces/evaluate-metric/rouge) 67 | - [11](https://github.com/neural-dialogue-metrics/rouge) 68 | - [12](https://paulminogue.com/posts/0de56ac8-914a-4056-9bf9-005572959bb1) 69 | - [13](https://studymachinelearning.com/cosine-similarity-text-similarity-metric/) 70 | - [14](https://intellica-ai.medium.com/comparison-of-different-word-embeddings-on-text-similarity-a-use-case-in-nlp-e83e08469c1c) 71 | - [15](https://arxiv.org/abs/2205.05092) 72 | - [16](https://stats.stackexchange.com/questions/415382/evaluating-word-embeddings-expected-cosine-distance) 73 | -------------------------------------------------------------------------------- /NLP/Practical_text_to_SQL_for_data_analytics_by_linkedin.md: -------------------------------------------------------------------------------- 1 | تجربه Linkedin از Text-to-SQL 2 | 3 | مساله اینه که خیلی از شرکت‌ها برای خودشون حجم زیادی داده دارند. داده‌هایی که در قالب یک سری جدول که خودشون شامل یک‌سری فیلد هستند ذخیره شده. کوئری زدن روی این داده‌ها حالا یکی از نیازمندی‌های پرتکرار و مهم برای هر شرکتی و کارمندانشه. نیازمندی که معمولا بر عهده data analyst هاست و برای حل اون باید یک پرسشی که در قالب زبان طبیعی هست رو به یک کوئری SQL ای تبدیل کنن. به این تسک در لیترچر هوش مصنوعی، Text to SQL گفته می‌شه. حالا لینکدین طی بلاگی نوشته که چه طور یک محصول AI ای برای Text to SQL برای شرکت خودشون درست کردند. مساله‌شون رو به صورت کلی این شکلی توضیح دادن که یک سری تیبل و دیتا داریم می‌خوایم یک سری سوال بپرسیم که پاسخش، کوئری روی این داده‌هاست. مسیری که ازش باید بگذریم اینه که با توجه به پرسش مطرح شده، جدول‌های مربوط بهش رو پیدا کنیم، روشون کوئری بزنیم و در صورت نیاز اصلاحشون کنیم و همچنین حواسمون باشه که یوزری که داره اون کوئری رو می‌زنه پرمیشن دسترسی به اون جدول‌ها رو داشته باشه. 4 | 5 | حالا نحوه حل کردن مساله از RAG می‌گذره و اینها هم به صورت RAG فرمولش کردند. برای این که بتونن RAG بزنن اولا تیم‌ها رو مجبور کردند که جدول‌های مهم و اسکیماشون رو مشخص کنند و برای تیبل‌ها و فیلدها توضیحات بنویسند. بعد این توضیحات رو دادن خود AI هم بر حسب داکیومنتیشن‌های موجود و صحبت‌های تو اسلک تکمیلشون کنه. بعدش هم EBR زدند و ۲۰ تا جدول رو به عنوان مرتبط ترین‌ها با کوئری برگردوندند. 6 | 7 | در مرحله بعدی حالا اومدن از خروجی مرحله قبلی استفاده کردند و سعی کردند تیبل‌های محدودتر ولی درست‌تری رو فیلتر کنند. برای این کار اولا اومدند از روی یوزر‌ها و تیبل‌ها و فیلد‌ها نالج گراف ساختند. یوزر این که به چه تیبل‌هایی دسترسی داره و تاریخچه تیبل گردیش چه شکلی بوده و تیبل‌ها هم این که هر تیبل چه فیچر‌هایی داره و یک سری کوئری مرتبط مثالی باهاش و فیلدها هم این که هر فیلد چه توضیحاتی داره و مثلا مقادیر top-k اش چه بودند و اینها. من به شخصه خیلی از این قسمت کارشون خوشم اومد. به نظرم جالب بود. بعد کوئری رو به همراه این نالچ گراف و ۲۰ تا تیبلی که فاز قبلی درآورده بودند دادند دست LLM و ازش خواستند ۷ تا تیبل مرتبط‌تر رو رنک کرد. در مرحله بعدی هم اومدند همین بلا رو سر فیلد‌های این ۷ تیبل درآوردند و فیلدها رو هم محدود کردند. 8 | 9 | حالا موقع ساختن کوئریه. برای این کار یک فرآیند iterative داشتند. به این صورت که LLM میاد از روی متن پرسش و تیبل‌ها و فیلد‌ها، یک پلن می‌نویسه و بعد برای هر مرحله از پلنش، گام به گام یک کوئری می‌نویسه (یعنی CoT زده). بعدش هر مرحله رو ران می‌کنن و خروجی رو براش یک مجموعه validator گذاشتند تا چک کنه آیا این کوئری درست و مربوط هست یا نه. در اینجا باز یک LLM دیگه گذاشتند که به نتایج و کوئری نگاه می‌کنه و در صورتی که تشخیص بده اشتباهی رخ داده، اون رو ارزیابی می‌کنه و درستش می‌کنه. (به حجم مهندسی کار دقت کنید). 10 | 11 | این جا تقریبا راه‌حل کلی تمومه. اومدند راجع به این توضیح دادند که روی user experience کار خیلی دقت و تمرکز گذاشتند و تجربه کاربری رو خیلی راحت کردند تا ملت بتونن ازش استفاده کنند. نسخه اولیه‌ محصولشون فقط کوئری SQL می‌نوشته، بعد فهمیدن که کاربرانشون ممکنه نیازهای دیگه‌ای هم داشته باشند، مثلا بخواد راجع به خود table و این که چی هست سوال بپرسه. برای همین یک ماژول intent detection هم باز طراحی کردند و گذاشتند که وظیفه‌اش تشخیص intent کاربر و هدایت اون به مسیر درسته. 12 | 13 | در اخر اذعان کردند که این راه حل خیلی هایپرپارامتر داره و قطعا نیاز هست که بنچمارک و کنترل بشه. برای این کار یک سری معیار تعریف کردند نظیر ریکال تیبل و فیلد، میزان هالوسینیشن مدل روی تیبل‌ها و فیلدها، ارورر سینتکسی، لیتنسی و ... و مدل رو کنترل کردند. برای جاج مدل هم از عوامل انسانی و باز یک LLM در حکم داور استفاده کردند. این شکلی که با یک عده ground truth اولیه شروع کردند و جواب مدل‌ رو چک ‌می‌کنند. این وسط فهمیدند که یک پرسش ممکنه چندین نوع راه حل و جواب داشته باشه و به کمک همین عوامل انسانی هر سری راه‌حل‌های درست برای یک پرسش رو به مجموعه ground truth شون اضافه می‌کنند و این جوری در طول زمان یک مجموعه برای کنترل کیفیت محصولشون داره. 14 | 15 | درسی که می‌گیریم اینه که ما می‌گیم RAG ولی در عمل هزار پیچش و نکته و ریزه‌کاری مهندسی هست که از RAG یک محصول خفن می‌سازه. فلذا اگر روی مساله‌ای دارید کار می‌کنید و سعی دارید با llm حلش کنید ولی به جواب جذابی نمی‌رسید شاید خیلی ساده‌ دارید مدلش می‌کنید. 16 | 17 | لینک بلاگ اصلی: 18 | https://www.linkedin.com/blog/engineering/ai/practical-text-to-sql-for-data-analytics 19 | 20 | لینک این مطلب (کانال تلگرام https://t.me/out_of_distribution): 21 | https://t.me/out_of_distribution/1122 22 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Great-Deep-Learning-Tutorials 2 | A Great Collection of Deep Learning Tutorials and Repositories 3 | 4 | ## General Deep Learning Tutorials: 5 | - [Browse state-of-the-art Deep Learning based Papers with their associated codes](https://paperswithcode.com/sota) [_Extremely Fantastic_] 6 | - [Deep-Learning-Roadmap](https://github.com/astorfi/Deep-Learning-Roadmap) 7 | - [DeepLizard](https://deeplizard.com/) [_Good Tutorials for Deep Learning_] 8 | - [Sebastian Ruder - Blog](https://ruder.io/) [_Great NLP & Deep Learning Posts_] 9 | - [Jeremy Jordan - Blog](https://www.jeremyjordan.me/author/jeremy/) 10 | - [Excellent Blog](https://lilianweng.github.io/lil-log/) 11 | - [Torchvision Release Notes](https://github.com/pytorch/vision/releases) [_Important_] 12 | - [The 6 most useful Machine Learning projects of the past year (2018)](https://towardsdatascience.com/the-10-most-useful-machine-learning-projects-of-the-past-year-2018-5378bbd4919f) 13 | - [ResNet Review](https://towardsdatascience.com/review-resnet-winner-of-ilsvrc-2015-image-classification-localization-detection-e39402bfa5d8) 14 | - [Receptive Field Estimation](https://github.com/fornaxai/receptivefield) [_Great_] 15 | - [An overview of gradient descent optimization algorithms](https://ruder.io/optimizing-gradient-descent/) [_Useful_] 16 | - [How to decide on learning rate](https://towardsdatascience.com/how-to-decide-on-learning-rate-6b6996510c98) 17 | - [Overview of State-of-the-art Machine Learning Algorithms per Discipline per Task](https://towardsdatascience.com/overview-state-of-the-art-machine-learning-algorithms-per-discipline-per-task-c1a16a66b8bb) 18 | - [Practical Machine Learning](https://github.com/youssefHosni/Practical-Machine-Learning) 19 | - [Awesome Machine Learning and AI Courses](https://github.com/luspr/awesome-ml-courses) 20 | - [UVA Deep Learning II Course](https://uvadl2c.github.io/) 21 | - [PyTorch Book](https://github.com/chenyuntc/pytorch-book) 22 | - [Fast.ai Course: Practical Deep Learning for Coders](https://course.fast.ai/Lessons/lesson1.html) [**Great**] 23 | - [Neuromatch Deep Learning Course](https://deeplearning.neuromatch.io/tutorials/intro.html) [**Great**] 24 | - [labmlai: 59 Implementations/tutorials of deep learning papers with side-by-side notes](https://github.com/labmlai/annotated_deep_learning_paper_implementations) [**Great**] 25 | - [labml.ai](https://nn.labml.ai/index.html) 26 | - [FightingCV-Paper-Reading: understand the most advanced research work in an easier way](https://github.com/xmu-xiaoma666/FightingCV-Paper-Reading) 27 | - [Learn PyTorch for Deep Learning: Zero to Mastery Course](https://github.com/mrdbourke/pytorch-deep-learning) [**Excellent**] 28 | - [ML Papers Explained](https://github.com/dair-ai/ML-Papers-Explained) [**Excellent**] 29 | - [Alpha Signal: Latest Research in Machine Learning](https://alphasignal.ai/) 30 | - [Harvard CS197: AI Research Experiences - The Course Book](https://docs.google.com/document/u/0/d/1uvAbEhbgS_M-uDMTzmOWRlYxqCkogKRXdbKYYT98ooc/mobilebasic#heading=h.bko37p9m9o8g) [**Excellent**] 31 | - [Deep learning jupyter notebook book](https://udlbook.github.io/udlbook/) 32 | - [A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT](https://arxiv.org/abs/2302.09419) 33 | - [interconnects.ai: Great AI Blog Posts & Podcasts](https://www.interconnects.ai/) 34 | - [The Fundamental of Modern Deep Learning with PyTorch (short Course)](https://www.linkedin.com/posts/sebastianraschka_github-rasbtpycon2024-tutorial-materials-activity-7196468139289677827-qUlf?utm_source=share&utm_medium=member_android) 35 | - [Google ML Crash Course](https://www.linkedin.com/posts/neilhoyne_google-lifeatgoogle-ai-activity-7262238201635868675-va5Q?utm_source=share&utm_medium=member_desktop) 36 | - [Intel AI Developer Course](https://www.linkedin.com/posts/eric-vyacheslav-156273169_want-to-become-a-top-ai-developer-intel-activity-7264293889837051904-nS1t?utm_source=share&utm_medium=member_desktop) 37 | - [EfficientML Course](https://www.youtube.com/playlist?list=PL80kAHvQbh-pT4lCkDT53zT8DKmhE0idB) [Great] 38 | - [Andrej Karpathy's Neural Networks: Zero to Hero Course](https://www.youtube.com/playlist?list=PLAqhIrjkxbuWI23v9cThsA9GvCAUhRvKZ) 39 | 40 | ## Deep Learning Useful Resources for Computer Vision: 41 | - [Great Deep Learning Resources for Computer Vision Tasks](https://github.com/ahkarami/Great-Deep-Learning-Tutorials/blob/master/ComputerVision.md) [_Excellent_] 42 | 43 | ## Deep Learning Useful Resources for Natural Language Processing (NLP): 44 | - [Great Deep Learning Resources for NLP Tasks](https://github.com/ahkarami/Great-Deep-Learning-Tutorials/blob/master/NLP.md) [_Excellent_] 45 | 46 | ## Deep Learning Useful Resources for Spoken Language Processing (Speech Processing): 47 | - [Great Deep Learning Resources for Speech Processing Tasks](https://github.com/ahkarami/Great-Deep-Learning-Tutorials/blob/master/Speech.md) [_Excellent_] 48 | 49 | ## Deep Learning & Machine Learning Useful Resources for General Data Science Tasks: 50 | - [Great Deep Learning Resources for Data Science Tasks](https://github.com/ahkarami/Great-Deep-Learning-Tutorials/blob/master/DataScience.md) [_Excellent_] 51 | 52 | ## General Notes about Generative AI: 53 | - [Generative AI in action: real-world applications and examples](https://lablab.ai/blog/generative-ai-in-action-real-world-applications-and-examples) 54 | 55 | ## Quantization & Distillation of Deep Learning Models: 56 | - [Quantization](https://nervanasystems.github.io/distiller/quantization/) 57 | - [Neural Network Distiller](https://github.com/NervanaSystems/distiller/) 58 | - [Introduction to Quantization on PyTorch](https://pytorch.org/blog/introduction-to-quantization-on-pytorch/) [_Excellent_] 59 | - [Dynamic Quantization in PyTorch](https://pytorch.org/tutorials/advanced/dynamic_quantization_tutorial.html) 60 | - [Static Quantization in PyTorch](https://pytorch.org/tutorials/advanced/static_quantization_tutorial.html) 61 | - [Intel(R) Math Kernel Library - Intel MKL-DNN](https://github.com/intel/mkl-dnn) 62 | - [Intel MKL-Dnn](https://01.org/mkl-dnn) 63 | - [ONNX Float32 to Float16](https://github.com/onnx/onnx-docker/blob/master/onnx-ecosystem/converter_scripts/float32_float16_onnx.ipynb) 64 | - [Neural Network Quantization Introduction](https://jackwish.net/neural-network-quantization-introduction.html) [_Tutorial_] 65 | - [Quantization in Deep Learning](https://medium.com/@joel_34050/quantization-in-deep-learning-478417eab72b) [_Tutorial_] 66 | - [Speeding up Deep Learning with Quantization](https://towardsdatascience.com/speeding-up-deep-learning-with-quantization-3fe3538cbb9) [_Tutorial_] 67 | - [Knowledge Distillation in Deep Learning](https://medium.com/analytics-vidhya/knowledge-distillation-dark-knowledge-of-neural-network-9c1dfb418e6a) 68 | - [Model Distillation Techniques for Deep Learning](https://heartbeat.fritz.ai/research-guide-model-distillation-techniques-for-deep-learning-4a100801c0eb) 69 | - [MMRazor: model compression toolkit](https://github.com/open-mmlab/mmrazor) [Great] 70 | - [FP8 Quantization: The Power of the Exponent](https://github.com/Qualcomm-AI-research/FP8-quantization) 71 | - [Quanto: a pytorch quantization toolkit](https://huggingface.co/blog/quanto-introduction) [**Great**] 72 | 73 | ## AutoML: 74 | - [Auto Gluon AI](https://auto.gluon.ai/stable/index.html#) 75 | - [AWS Auto Gluon](https://github.com/awslabs/autogluon) 76 | 77 | ## Diffusion Models: 78 | - [Diffusion Models via lilianweng](https://lilianweng.github.io/posts/2021-07-11-diffusion-models/) 79 | - [Diffusion Models Papers Survey Taxonomy](https://github.com/YangLing0818/Diffusion-Models-Papers-Survey-Taxonomy) 80 | - [Phenaki: a text-to-video model](https://github.com/LAION-AI/phenaki) 81 | 82 | ## Multimodal Deep Learning: 83 | - [Multimodal Deep Learning Book](https://arxiv.org/abs/2301.04856) 84 | - [Understanding MultiModal LLMs](https://www.linkedin.com/posts/sebastianraschka_there-has-been-a-lot-of-new-research-on-the-activity-7258836067129139200-QJkr?utm_source=share&utm_medium=member_desktop) 85 | 86 | ## Deep Reasoning: 87 | - [What’s Next For AI? Enter: Deep Reasoning](https://towardsdatascience.com/whats-next-for-ai-enter-deep-reasoning-fae8b131962a) 88 | - [Deep Learning approaches to understand Human Reasoning](https://towardsdatascience.com/deep-learning-approaches-to-understand-human-reasoning-46f1805d454d) 89 | 90 | ## Deep Reinforcement Learning (Great Courses & Tutorials): 91 | - [A Free course in Deep Reinforcement Learning from beginner to expert](https://simoninithomas.github.io/Deep_reinforcement_learning_Course/) [_Great_] 92 | - [Deep Reinforcement Learning Algorithms with PyTorch](https://github.com/p-christ/Deep-Reinforcement-Learning-Algorithms-with-PyTorch) 93 | - [Deep Reinforcement Learning - CS 285 Berkeley Course](rail.eecs.berkeley.edu/deeprlcourse/) 94 | - [solutions to UC Berkeley CS 285](https://github.com/xuanlinli17/CS285_Fa19_Deep_Reinforcement_Learning) 95 | - [Reinforcement Learning: An Introduction - main book in this field](http://www.incompleteideas.net/book/the-book-2nd.html) 96 | - [CS234: Reinforcement Learning Course](https://www.youtube.com/playlist?list=PLoROMvodv4rOSOPzutgyCTapiGlY2Nd8u) 97 | - [Introduction to Reinforcement Learning Course - by DeepMind](https://www.youtube.com/playlist?list=PLoROMvodv4rOSOPzutgyCTapiGlY2Nd8u) 98 | 99 | ## Graph Neural Networks: 100 | - [An Introduction to Graph Neural Networks](https://towardsdatascience.com/an-introduction-to-graph-neural-networks-e23dc7bdfba5) 101 | - [How to Train Graph Convolutional Network Models in a Graph Database](https://towardsdatascience.com/how-to-train-graph-convolutional-network-models-in-a-graph-database-5c919a2f95d7) 102 | - [A comprehensive survey on graph neural networks](https://arxiv.org/pdf/1901.00596) 103 | - [Graph Neural Networks: A Review of Methods and Applications](https://arxiv.org/abs/1812.08434) 104 | 105 | ### Graph Neural Networks Frameworks: 106 | - [Spektral](https://github.com/danielegrattarola/spektral) 107 | - [Deep Graph Library - DGL](https://www.dgl.ai/) 108 | - [PyTorch Geometric - PyG](https://github.com/rusty1s/pytorch_geometric) 109 | - [ptgnn: A PyTorch GNN Library](https://github.com/microsoft/ptgnn) 110 | - [Graph Data Augmentation Papers](https://github.com/zhao-tong/graph-data-augmentation-papers) 111 | - [Neo4j: Graph Data Platform](https://neo4j.com/) 112 | 113 | ## Best Practices for Training Deep Models: 114 | 115 | ### General Notes for Training Deep Models: 116 | - [Deep Learning Tuning Playbook](https://github.com/google-research/tuning_playbook) 117 | 118 | ### PyTorch Lightening Notes & Accumulate Gradients: 119 | - [PyTorch Lightening: Effective Training Techniques](https://pytorch-lightning.readthedocs.io/en/latest/advanced/training_tricks.html) 120 | - [Gradient Accumulation in PyTorch](https://kozodoi.me/python/deep%20learning/pytorch/tutorial/2021/02/19/gradient-accumulation.html) 121 | 122 | ### Loss Functions: 123 | - [Loss Functions Explained](https://medium.com/deep-learning-demystified/loss-functions-explained-3098e8ff2b27) 124 | 125 | ### Imbalanced Dataset Handling: 126 | - [deal with an imbalanced dataset using weightedrandomsampler](https://androidkt.com/deal-with-an-imbalanced-dataset-using-weightedrandomsampler-in-pytorch/) 127 | - [imbalanced-dataset-sampler](https://github.com/ufoym/imbalanced-dataset-sampler) [Great] 128 | - [demystifying pytorchs weightedrandomsampler](https://towardsdatascience.com/demystifying-pytorchs-weightedrandomsampler-by-example-a68aceccb452) 129 | - [weighted random sampler oversample or undersample](https://stackoverflow.com/questions/67799246/weighted-random-sampler-oversample-or-undersample) 130 | 131 | ### Weight Initialization: 132 | - [Deep Learning Best Practices (1) - Weight Initialization](https://medium.com/usf-msds/deep-learning-best-practices-1-weight-initialization-14e5c0295b94) 133 | 134 | ### Batch Normalization: 135 | - [Batch Normalization in Neural Networks](https://towardsdatascience.com/batch-normalization-in-neural-networks-1ac91516821c) 136 | - [Batch Normalization and Dropout in Neural Networks](https://towardsdatascience.com/batch-normalization-and-dropout-in-neural-networks-explained-with-pytorch-47d7a8459bcd) 137 | - [Difference between Local Response Normalization and Batch Normalization](https://towardsdatascience.com/difference-between-local-response-normalization-and-batch-normalization-272308c034ac) 138 | 139 | ### Learning Rate Scheduling & Initialization: 140 | - [Automated Learning Rate Suggester](https://forums.fast.ai/t/automated-learning-rate-suggester/44199) 141 | - [Learning Rate Finder - fastai](https://fastai1.fast.ai/callbacks.lr_finder.html) 142 | - [Cyclical Learning Rates for Training Neural Networks](https://arxiv.org/abs/1506.01186) 143 | - [ignite - Example of FastaiLRFinder](https://github.com/pytorch/ignite/blob/master/examples/notebooks/FastaiLRFinder_MNIST.ipynb) 144 | - [Find Learning Rate - a gist code](https://gist.github.com/colllin/738cd2a9f0abec9be5e8b9becc23a812) 145 | - [Learning rate finder - PyTorch Lightning](https://pytorch-lightning.readthedocs.io/en/1.1.3/lr_finder.html) 146 | - [RAdam - On the Variance of the Adaptive Learning Rate and Beyond](https://github.com/LiyuanLucasLiu/RAdam) 147 | 148 | ### Early Stopping: 149 | - [Early Stopping in PyTorch - Bjarten](https://github.com/Bjarten/early-stopping-pytorch) 150 | - [Catalyst - Early Stopping](https://catalyst-team.github.io/catalyst/faq/early_stopping.html) 151 | - [ignite - Early Stopping](https://github.com/pytorch/ignite/blob/master/ignite/handlers/early_stopping.py) 152 | - [PyTorch High-Level Training Sample](https://github.com/ncullen93/torchsample/blob/master/README.md) 153 | - [PyTorch Discussion about Early Stopping](https://discuss.pytorch.org/t/early-stopping-in-pytorch/18800) 154 | 155 | ### Tuning Guide Recipes: 156 | - [PyTorch Tuning Guide Tutorial](https://pytorch.org/tutorials/recipes/recipes/tuning_guide.html) 157 | - [PyTorch memory leak with dynamic size tensor input](https://github.com/pytorch/pytorch/issues/29893) 158 | - [Karpathy: A Recipe for Training Neural Networks](http://karpathy.github.io/2019/04/25/recipe/) 159 | 160 | ### Training Optimizer: 161 | - [What is gradient accumulation in deep learning](https://towardsdatascience.com/what-is-gradient-accumulation-in-deep-learning-ec034122cfa) 162 | 163 | ### PyTorch running & training on TPU (colab): 164 | - [PyTorch XLA](https://github.com/pytorch/xla) 165 | - [PyTorch XLA Colab](https://github.com/pytorch/xla/tree/master/contrib/colab) 166 | 167 | ### Evaluation Metrics: 168 | - [Performance Metrics for Classification Problems in ML](https://medium.com/@MohammedS/performance-metrics-for-classification-problems-in-machine-learning-part-i-b085d432082b) 169 | 170 | ### Validating ML Models: 171 | - [Deepchecks: Validating ML Models & Data](https://github.com/deepchecks/deepchecks) 172 | 173 | ### Optimizing models when run on GPU: 174 | - [Tips for reducing vram of gpu memories](https://www.linkedin.com/posts/pauliusztin_machinelearning-mlops-datascience-activity-7137704771905277953-G8Qt?utm_source=share&utm_medium=member_desktop) 175 | 176 | ## Conferences News: 177 | - [Latest Computer Vision Trends from CVPR 2019](https://towardsdatascience.com/latest-computer-vision-trends-from-cvpr-2019-c07806dd570b) 178 | - [Interesting 2019 CVPR papers](https://medium.com/@mattmiesnieks/interesting-2019-cvpr-papers-865e303db5ca) 179 | - [Summaries of CVPR papers on ShortScience.org](https://www.shortscience.org/venue?key=conf/cvpr) 180 | - [Summaries of ICCV papers on ShortScience.org](https://www.shortscience.org/venue?key=conf/iccv) 181 | - [Summaries of ECCV papers on ShortScience.org](https://www.shortscience.org/venue?key=conf/eccv) 182 | - [Meta ICLR 2024 Top Papers](https://www.linkedin.com/posts/aiatmeta_iclr2024-activity-7194398361943171074-XiVG?utm_source=share&utm_medium=member_android) 183 | 184 | ## Deep Learning Frameworks and Infrustructures: 185 | - [set-up a Paperspace GPU Server](https://towardsdatascience.com/how-to-set-up-a-powerful-and-cost-efficient-gpu-server-for-deep-learning-aa1de0d4ea56) 186 | - [Distributed ML with OpenMPI](https://clusterone.com/tutorials/openmpi-introduction) 187 | - [Tensorflow 2.0 vs Mxnet](https://medium.com/@mouryarishik/tensorflow-2-0-vs-mxnet-41edd3b7574f) 188 | - [TensorFlow is dead, long live TensorFlow!](https://hackernoon.com/tensorflow-is-dead-long-live-tensorflow-49d3e975cf04) 189 | 190 | ## Great Libraries: 191 | - [The Unified Machine Learning Framework](https://github.com/unifyai/ivy) 192 | - [Skorch - A scikit-learn compatible neural network library that wraps PyTorch](https://github.com/skorch-dev/skorch) 193 | - [Hummingbird - traditional ML models into tensor computations via PyTorch](https://github.com/microsoft/hummingbird) 194 | - [BoTorch - Bayesian Optimization in PyTorch](https://botorch.org/) 195 | - [torchvision 0.3: segmentation, detection models, new datasets and more](https://pytorch.org/blog/torchvision03/) 196 | - [TorchAudio: an audio library for PyTorch](https://github.com/pytorch/audio) 197 | - [AudTorch](https://github.com/audeering/audtorch) 198 | - [TorchAudio-Contrib](https://github.com/keunwoochoi/torchaudio-contrib) 199 | - [fastText - Facebook AI Research (FAIR)](https://fasttext.cc/) 200 | - [Fairseq - Facebook AI Research (FAIR)](https://github.com/pytorch/fairseq) 201 | - [ParlAI - dialogue models - Facebook AI Research (FAIR)](https://parl.ai/) 202 | - [DALI - highly optimized engine for data pre-processing](https://github.com/NVIDIA/DALI) 203 | - [Netron - GitHub](https://github.com/lutzroeder/netron) [_Visualizer for deep learning Models (Excellent)_] 204 | - [Netron - Web Site](https://www.lutzroeder.com/ai) 205 | - [JupyterLab GPU Dashboards](https://github.com/rapidsai/jupyterlab-nvdashboard) [_Good_] 206 | - [PyTorch Hub](https://pytorch.org/hub) 207 | - [Neural Structured Learning (NSL) in TensorFlow](https://github.com/tensorflow/neural-structured-learning) 208 | - [Pywick - High-Level Training framework for Pytorch](https://github.com/achaiah/pywick) 209 | - [torchbearer: A model fitting library for PyTorch](https://github.com/pytorchbearer/torchbearer) 210 | - [torchlayers - Shape inference for PyTorch (like in Keras)](https://github.com/szymonmaszke/torchlayers) 211 | - [torchtext - GitHub](https://github.com/pytorch/text) 212 | - [torchtext - Doc](https://torchtext.readthedocs.io/en/latest/) 213 | - [Optuna - hyperparameter optimization framework](https://optuna.org/) 214 | - [PyTorchLightning](https://github.com/PyTorchLightning/pytorch-lightning) 215 | - [Nvidia - runx - An experiment management tool](https://github.com/NVIDIA/runx) 216 | - [MLogger: a Machine Learning logger](https://github.com/oval-group/mlogger) 217 | - [ClearML - ML/DL development and production suite](https://github.com/allegroai/clearml) 218 | - [Lime: Explaining the predictions of any ML classifier](https://github.com/marcotcr/lime) 219 | - [Microsoft UniLM AI](https://github.com/microsoft/unilm) [Great] 220 | - [mlnotify: No need to keep checking your training](https://github.com/aporia-ai/mlnotify) 221 | - [NVIDIA NeMo - toolkit for creating Conversational AI (ASR, TTS, and NLP)](https://github.com/NVIDIA/NeMo) 222 | - [Microsoft DeepSpeed](https://github.com/microsoft/DeepSpeed) 223 | - [Mojo: a new programming language for AI developers](https://www.modular.com/mojo) 224 | - [MLX: An array framework for Apple silicon](https://github.com/ml-explore/mlx) 225 | 226 | ## Great Models: 227 | - [ResNext WSL](https://pytorch.org/hub/facebookresearch_WSL-Images_resnext/) [_Great Pretrained Model_] 228 | - [Semi-Weakly Supervised (SWSL) ImageNet Models](https://pytorch.org/hub/facebookresearch_semi-supervised-ImageNet1K-models_resnext/) [_Great Pretrained Model_] 229 | - [Deep High-Resolution Representation Learning (HRNet)](https://jingdongwang2017.github.io/Projects/HRNet/) 230 | 231 | ## Deep Model Conversion: 232 | - [Convert Full ImageNet Pre-trained Model from MXNet to PyTorch](https://blog.paperspace.com/convert-full-imagenet-pre-trained-model-from-mxnet-to-pytorch/) [_Great_] 233 | - [ONNX Runtime](https://github.com/microsoft/onnxruntime) 234 | 235 | ## Great Deep Learning Repositories (for learning DL-based programming): 236 | - [deeplearning-models - PyTorch & TensorFlow Learning](https://github.com/rasbt/deeplearning-models) [_Very Excellent Repository_] 237 | - [PyTorch Image Models](https://github.com/rwightman/pytorch-image-models) [_Great_] 238 | - [5 Advanced PyTorch Tools to Level up Your Workflow](https://towardsdatascience.com/5-advanced-pytorch-tools-to-level-up-your-workflow-d0bcf0603ad5) [_Interesting_] 239 | 240 | ## PyTorch High-Level Libraries: 241 | - [Catalyst - PyTorch framework for Deep Learning research and development](https://github.com/catalyst-team/catalyst) [_Great_] 242 | - [PyTorch Lightning - GitHub](https://github.com/PyTorchLightning/pytorch-lightning) [_Great_] 243 | - [PyTorch Lightning - Web Page](https://pytorchlightning.ai/) 244 | - [Ignite - GitHub](https://github.com/pytorch/ignite) [_Great_] 245 | - [Ignite - Web Page](https://pytorch.org/ignite/) 246 | - [TorchMetrics](https://torchmetrics.readthedocs.io/en/latest/) 247 | - [Ludwig AI: Data-centric declarative deep learning framework](https://github.com/ludwig-ai/ludwig) [**Great**] 248 | - [PyTorch Kineto: CPU+GPU Profiling library](https://github.com/pytorch/kineto/) 249 | - [PyTorch Profiler](https://pytorch.org/docs/master/profiler.html) 250 | - [PyTorch Benchmarks](https://github.com/pytorch/benchmark) 251 | 252 | ## Annotation Tools: 253 | - [label-studio](https://github.com/heartexlabs/label-studio) 254 | - [label-studio with RTL Support (for Persian)](https://github.com/mmaghajani/label-studio) 255 | 256 | ## Other: 257 | - [Clova AI Research - NAVER & LINE](https://github.com/clovaai) 258 | - [Exploring Weight Agnostic Neural Networks](https://ai.googleblog.com/2019/08/exploring-weight-agnostic-neural.html) 259 | - [Weight Agnostic Neural Networks](https://weightagnostic.github.io/) 260 | - [Weight Agnostic Neural Networks - GitHub](https://github.com/google/brain-tokyo-workshop/tree/master/WANNRelease) 261 | - [SAM: Sharpness-Aware Minimization for Efficiently Improving Generalization](https://github.com/google-research/sam) 262 | - [Qualcomm Discusses Secret Dataset Generation Data](https://www.qualcomm.com/news/onq/2021/09/16/qa-ai-researcher-roland-memisevic-discusses-secret-dataset-generation-data) 263 | - [State of AI Report 2021](https://www.stateof.ai/) 264 | - [State of AI Report 2024](https://www.stateof.ai/) 265 | - [Project Blink: AI-powered video editing on the web](https://labs.adobe.com/projects/blink/) 266 | - [PyTorch Incremental Learning](https://github.com/yaoyao-liu/class-incremental-learning) 267 | - [Google Research, 2022 & Beyond: Language, Vision and Generative Models](https://ai.googleblog.com/2023/01/google-research-2022-beyond-language.html?m=1&fbclid=PAAabtVizCEKhFC2kttHKozuEz4FX1cphjNDQVVL-kFZHA11GP9AVJ6rl9W-k) 268 | - [Elicit: Ask a research question](https://elicit.org/) [Interesting] 269 | - [Google People + AI Research (PAIR)](https://pair.withgoogle.com/) [Interesting business based AI topics] 270 | - [Google Illuminate](https://illuminate.google.com/home) [Great] 271 | - [Google Learn About](https://learning.google.com/experiments/learn-about/signup) [Great] 272 | -------------------------------------------------------------------------------- /Speech.md: -------------------------------------------------------------------------------- 1 | # Great Deep Learning Tutorials & Resources for Speech Processing 2 | A Great Collection of Deep Learning Tutorials and Repositories for Speech Processing 3 | 4 | ## General (Spoken Language Processing (Speech Processing)): 5 | - [Audio Classification](https://towardsdatascience.com/audio-classification-using-fastai-and-on-the-fly-frequency-transforms-4dbe1b540f89) [_Great_] 6 | - [Building a Dead Simple Word Recognition Engine Using Convnet](https://blog.manash.me/building-a-dead-simple-word-recognition-engine-using-convnet-in-keras-25e72c19c12b) 7 | - [Identifying the Genre of a Song with Neural Networks](https://medium.com/@navdeepsingh_2336/identifying-the-genre-of-a-song-with-neural-networks-851db89c42f0) 8 | - [Modelling audio signal using visual features](https://raghavgoyal14.github.io/2018/04/12/audio-via-vid-features.html) 9 | - [ESC-50: Dataset for Environmental Sound Classification](https://github.com/karolpiczak/ESC-50) 10 | - [Kaldi Speech Recognition Toolkit](https://github.com/kaldi-asr/kaldi) 11 | - [PyTorch-Kaldi](https://github.com/mravanelli/pytorch-kaldi) 12 | - [SpeechBrain - PyTorch-based Speech Toolkit](https://speechbrain.github.io/) 13 | - [How to start with Kaldi and Speech Recognition](https://towardsdatascience.com/how-to-start-with-kaldi-and-speech-recognition-a9b7670ffff6) 14 | - [A 2019 Guide to Speech Synthesis with Deep Learning](https://heartbeat.fritz.ai/a-2019-guide-to-speech-synthesis-with-deep-learning-630afcafb9dd) 15 | - [A 2019 Guide for Automatic Speech Recognition](https://heartbeat.fritz.ai/a-2019-guide-for-automatic-speech-recognition-f1e1129a141c) 16 | - [PyKaldi](https://github.com/pykaldi/pykaldi) 17 | - [WaveNet vocoder](https://github.com/r9y9/wavenet_vocoder) 18 | - [nnAudio - audio processing toolbox using PyTorch](https://github.com/KinWaiCheuk/nnAudio) 19 | - [Athena - open-source implementation of end-to-end speech processing engine](https://github.com/athena-team/athena/tree/simclr) 20 | - [Pydub - manipulate audio](https://github.com/jiaaro/pydub) 21 | - [pyAcoustics - analyzing acoustics from audio files](https://github.com/timmahrt/pyAcoustics) 22 | - [ESPnet: end-to-end speech processing toolkit](https://github.com/espnet/espnet) 23 | - [WeNet](https://github.com/wenet-e2e/wenet) [Great] 24 | - [WeNet Android App](https://github.com/wenet-e2e/wenet/tree/main/runtime/device/android/wenet) 25 | - [K2: FSA/FST algorithms, differentiable, with PyTorch compatibility](https://github.com/k2-fsa/k2) 26 | - [Microsoft NeuralSpeech](https://github.com/microsoft/NeuralSpeech) 27 | - [Great Speech Tutorials: alphacephei](https://alphacephei.com/nsh/) 28 | - [AssemblyAI Lead Speech AI Models](https://www.assemblyai.com/products?utm_source=alphasignal&utm_medium=newsletter_sponsor&utm_campaign=spotlight&utm_content=82624) 29 | - [open-mmlab Amphion: An Open-Source Audio, Music, and Speech Generation Toolkit](https://github.com/open-mmlab/Amphion) 30 | - [HuggingFace Speech-to-Speech Library](https://github.com/huggingface/speech-to-speech) [**Great**] 31 | - [HuggingFace Speech-to-Speech Library News](https://www.linkedin.com/posts/andresmarafioti_introducing-hugging-faces-speech-to-speech-activity-7231548059388723201-1wIe?utm_source=share&utm_medium=member_desktop) 32 | - [NeMo - toolkit for Conversational AI](https://github.com/NVIDIA/NeMo) [_Excellent_] 33 | 34 | ## Text to Speech (TTS): 35 | - [Glow-TTS](https://github.com/jaywalnut310/glow-tts) 36 | - [ForwardTacotron](https://github.com/as-ideas/ForwardTacotron) 37 | - [WaveRNN Vocoder + TTS](https://github.com/fatchord/WaveRNN) 38 | - [Deep Voice 3 PyTorch](https://github.com/r9y9/deepvoice3_pytorch) 39 | - [MelGAN - TTS - version1](https://github.com/descriptinc/melgan-neurips) 40 | - [MelGAN - TTS - version2](https://github.com/seungwonpark/melgan) 41 | - [FastSpeech - TTS - version1](https://github.com/xcmyz/FastSpeech) 42 | - [FastSpeech - TTS - version2](https://github.com/ming024/FastSpeech2) 43 | - [Speedy Speech](https://github.com/janvainer/speedyspeech) 44 | - [Mozilla - TTS](https://github.com/mozilla/TTS) 45 | - [YourTTS: Zero-Shot Multi-Speaker TTS](https://github.com/edresson/yourtts) 46 | - [YourTTS: Zero-Shot Multi-Speaker Text Synthesis and Voice Conversion](https://coqui.ai/blog/tts/yourtts-zero-shot-text-synthesis-low-resource-languages/) 47 | - [Nix-TTS](https://github.com/rendchevi/nix-tts) 48 | - [TorToiSe](https://github.com/neonbjb/tortoise-tts) 49 | - [Amazon TTS Group's Research](https://www.amazon.science/blog/amazon-text-to-speech-groups-research-at-icassp-2022) 50 | - [NVIDIA RADTTS](https://github.com/NVIDIA/radtts) 51 | - [CanTTS: a single-speaker Cantonese speech dataset for TTS](https://github.com/parami-ai/CanTTS) 52 | - [Lightning Fast Speech2](https://github.com/MiniXC/LightningFastSpeech2) 53 | - [ProDiff: Progressive Fast Diffusion Model For High-Quality TTS](https://github.com/Rongjiehuang/ProDiff) 54 | - [TF light model (Mozilla tacotron2)](https://github.com/mozilla/TTS/blob/master/notebooks/DDC_TTS_and_MultiBand_MelGAN_TFLite_Example.ipynb) 55 | - [Lightweight end-to-end TTS](https://github.com/MasayaKawamura/MB-iSTFT-VITS) 56 | - [SiFiGAN](https://github.com/chomeyama/SiFiGAN) 57 | - [Neon TTS Plugin Coqui](https://huggingface.co/spaces/neongeckocom/neon-tts-plugin-coqui) 58 | - [VocBench: A Neural Vocoder Benchmark for Speech Synthesis](https://github.com/facebookresearch/vocoder-benchmark) 59 | - [MQTTS: Quantized Approach for Text to Speech Synthesis](https://github.com/b04901014/MQTTS) 60 | - [VITS Fast Fine-tuning: fast speaker adaptation TTS](https://github.com/Plachtaa/VITS-fast-fine-tuning) 61 | - [Larynx: A fast, local neural TTS](https://github.com/rhasspy/larynx2) 62 | - [BigVGAN: A Universal Neural Vocoder with Large-Scale Training](https://github.com/NVIDIA/BigVGAN) 63 | - [Bark: Text-Prompted Generative Audio Model](https://github.com/suno-ai/bark) 64 | - [FaceBook Massively Multilingual Speech (MMS)](https://github.com/facebookresearch/fairseq/tree/main/examples/mms) 65 | - [AudioLDM2: unified framework for text-to-audio generation](https://github.com/haoheliu/AudioLDM2) 66 | - [MetaVoice-1B: a 1.2B parameter base model trained on 100K hours of speech for TTS](https://github.com/metavoiceio/metavoice-src) 67 | - [Parler TTS](https://www.linkedin.com/posts/yoach-lacombe_introducing-data-speech-a-fully-open-source-activity-7183864961905872896-1EBe?utm_source=share&utm_medium=member_desktop) 68 | - [IMS-Toucan TTS: the first TTS System in over 7000 languages](https://github.com/DigitalPhonetics/IMS-Toucan) 69 | - [E2 TTS](https://www.linkedin.com/posts/naoyuki-kanda-16a00511b_e2-tts-microsoft-research-activity-7211927392804593665-ATUA?utm_source=share&utm_medium=member_android) 70 | - [Mars5 TTS](https://www.linkedin.com/posts/vaibhavs10_mars5-tts-open-source-text-to-speech-with-activity-7211102752305819650-RJwf?utm_source=share&utm_medium=member_desktop) 71 | - [Nvidia NeMo T5-TTS Model](https://www.linkedin.com/posts/subhankar-19_addressing-hallucinations-in-speech-synthesis-activity-7214352391993765889-EGVu?utm_source=share&utm_medium=member_desktop) 72 | - [Parler-TTS: fully open-source high-quality TTS](https://huggingface.co/collections/parler-tts/parler-tts-fully-open-source-high-quality-tts-66164ad285ba03e8ffde214c) 73 | - [Fish Speech TTS Models](https://github.com/fishaudio/fish-speech) 74 | - [Fish Speech V1.4: a leading text-to-speech (TTS) model](https://huggingface.co/fishaudio/fish-speech-1.4) 75 | - [FireRedTTS: A Foundation Text-To-Speech Framework for Industry-Level Generative Speech Applications](https://github.com/FireRedTeam/FireRedTTS) 76 | - [XTTS-v2](https://www.linkedin.com/posts/josh-r-meyer_its-been-almost-a-year-since-we-released-activity-7254933062004666368-FL6T?utm_source=share&utm_medium=member_desktop) 77 | - [Smoll TTS Models](https://www.linkedin.com/posts/vaibhavs10_smol-tts-models-are-here-outetts-01-350m-activity-7259247217750728704-OkZM?utm_source=share&utm_medium=member_desktop) 78 | 79 | ## Automatic Speech Recognition (ASR) & Speech to Text (STT): 80 | - [OpenSpeech](https://github.com/openspeech-team/openspeech) [Great] 81 | - [wav2letter++](https://github.com/facebookresearch/wav2letter) 82 | - [End-to-End ASR - PyTorch](https://github.com/Alexander-H-Liu/End-to-end-ASR-Pytorch) 83 | - [NeuralSP](https://github.com/hirofumi0810/neural_sp) 84 | - [Silero Speech-To-Text Models - PyTorch Hub](https://pytorch.org/hub/snakers4_silero-models_stt/) 85 | - [Silero Models - GitHub](https://github.com/snakers4/silero-models) 86 | - [Hugging Face’s Wav2Vec2 & its First ASR Model](https://www.analyticsvidhya.com/blog/2021/02/hugging-face-introduces-the-first-automatic-speech-recognition-model-wav2vec2/) 87 | - [Hugging Face - wav2vec2](https://huggingface.co/transformers/model_doc/wav2vec2.html) 88 | - [PyTorch Wav2Vec](https://github.com/pytorch/fairseq/tree/master/examples/wav2vec/unsupervised) 89 | - [Self-training and pre-training, understanding the wav2vec series](https://maelfabien.github.io/machinelearning/wav2vec/#) 90 | - [Conformer](https://github.com/sooftware/conformer) 91 | - [Emformer: RNNT Model](https://github.com/pytorch/audio/tree/main/examples/asr/emformer_rnnt) 92 | - [Emformer Paper](https://arxiv.org/abs/2010.10759) 93 | - [Nextformer](https://github.com/tuanio/nextformer) 94 | - [Keras based Training a CTC-based model for ASR](https://keras.io/examples/audio/ctc_asr/) 95 | - [alphacephei: citrinet](https://alphacephei.com/nsh/2021/04/23/citrinet.html) 96 | - [Coqui-ai STT](https://github.com/coqui-ai/STT) 97 | - [vosk Framework](https://alphacephei.com/vosk/) 98 | - [vosk Framework GitHub](https://github.com/alphacep/vosk-api) 99 | - [fairseq](https://github.com/pytorch/fairseq) 100 | - [TensorFlowASR](https://github.com/TensorSpeech/TensorFlowASR) [Good] 101 | - [Assembly AI ASR api](https://www.assemblyai.com/) 102 | - [Assembly AI: Building an End-to-End Speech Recognition Model in PyTorch](https://www.assemblyai.com/blog/end-to-end-speech-recognition-pytorch/) [Great] 103 | - [BigSSL: Large-Scale Semi-Supervised Learning for Automatic Speech Recognition](https://arxiv.org/pdf/2109.13226.pdf) 104 | - [Tencent AI Lab: 3M-ASR](https://github.com/tencent-ailab/3m-asr) 105 | - [wav2seq](https://github.com/asappresearch/wav2seq) 106 | - [WavPrompt: speech understanding that leveraging the few-shot learning](https://github.com/Hertin/WavPrompt) 107 | - [Recent Advances in End-to-End Automatic Speech Recognition](https://www.nowpublishers.com/article/Details/SIP-2021-0050) [Interesting Survey] 108 | - [SpeechT5](https://github.com/microsoft/SpeechT5) [Interesting] 109 | - [TransFusion: Transcribing Speech with Multinomial Diffusion](https://github.com/RF5/transfusion-asr) 110 | - [Alibaba FunASR](https://github.com/alibaba-damo-academy/FunASR) 111 | - [Openai whisper ASR Model](https://github.com/openai/whisper) [Interesting] 112 | - [Openai whisper ASR Model Blog](https://openai.com/blog/whisper/) 113 | - [Explanation of OpenAI whisper ASR Model](https://www.linkedin.com/posts/aladdin-persson-a95384153_a-couple-of-months-ago-openai-released-its-activity-7009429017552375808-6zCI/?utm_source=share&utm_medium=member_android) 114 | - [High-performance inference of Whisper ASR Model](https://github.com/Const-me/Whisper) 115 | - [Insanely fast whisper (very fast whisper)](https://www.linkedin.com/posts/liorsinclair_you-can-now-transcribe-25-hours-of-audio-activity-7136072189933408256-G3mL?utm_source=share&utm_medium=member_desktop) 116 | - [Google Universal Speech Model (USM)](https://ai.googleblog.com/2023/03/universal-speech-model-usm-state-of-art.html?m=1) 117 | - [FaceBook Massively Multilingual Speech (MMS)](https://github.com/facebookresearch/fairseq/tree/main/examples/mms) 118 | - [SeamlessM4T: Github](https://github.com/facebookresearch/seamless_communication) 119 | - [SeamlessM4T: Meta AI Blog](https://ai.meta.com/blog/seamless-m4t/) 120 | - [SeamlessM4T: Paper](https://dl.fbaipublicfiles.com/seamless/seamless_m4t_paper.pdf) 121 | - [SeamlessM4T: Demo](https://seamless.metademolab.com/demo) 122 | - [SeamlessM4T: HuggingFace Demo](https://huggingface.co/spaces/facebook/seamless_m4t) 123 | - [SeamlessM4T v2](https://www.linkedin.com/posts/aiatmeta_today-were-sharing-the-next-milestone-in-activity-7136104235166830592-7ROv?utm_source=share&utm_medium=member_desktop) 124 | - [WhisperFusion: Whisper + Mistral](https://github.com/collabora/WhisperFusion) 125 | - [NeMo Canary-1B ASR Model](https://huggingface.co/spaces/nvidia/canary-1b) 126 | - [NeMo Canary-1B Linkedin Post](https://www.linkedin.com/posts/stevehuanghe_nvidia-nemo-team-is-thrilled-to-announce-activity-7161407409448505344-HvrJ?utm_source=share&utm_medium=member_desktop) 127 | - [Google Chirp: Universal speech model (USM)](https://cloud.google.com/speech-to-text/v2/docs/chirp-model) [Great] 128 | - [Whisper V3 Turbo Model Linkedin Post](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_openai-has-released-new-whisper-models-activity-7246765913331367936-NP3N?utm_source=share&utm_medium=member_desktop) 129 | - [Gooya v1 Persian ASR Model](https://www.linkedin.com/posts/vargha-khallokhi-20b98316b_gooya-v1-persian-speech-recognition-a-hugging-activity-7258536721343016960-yfzN?utm_source=share&utm_medium=member_desktop) 130 | - [HuggingFace Open ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard) 131 | 132 | ## AudioLLM: 133 | - [Qwen-Audio](https://github.com/QwenLM/Qwen-Audio) 134 | - [Qwen2-Audio](https://github.com/QwenLM/Qwen2-Audio) 135 | - [Qwen2-Audio Blog](https://qwenlm.github.io/blog/qwen2-audio/) 136 | 137 | ## Speech to Speech Models: 138 | - [Speech To Speech: an effort for an open-sourced and modular GPT4-o](https://github.com/huggingface/speech-to-speech) 139 | - [Huggingface Multilingual Speech to Speech Library](https://www.linkedin.com/posts/andresmarafioti_introducing-hugging-faces-multilingual-activity-7236625231505731585-Pbps?utm_source=share&utm_medium=member_desktop) 140 | - [Moshi Speech to Speech Model](https://www.linkedin.com/posts/andresmarafioti_%F0%9D%97%95%F0%9D%97%B6%F0%9D%97%B4-%F0%9D%97%BB%F0%9D%97%B2%F0%9D%98%84%F0%9D%98%80-kyutai-just-open-sourced-activity-7242210751275311104-PQEd?utm_source=share&utm_medium=member_desktop) 141 | - [Moshi Speech to Speech Model - Link2](https://www.linkedin.com/posts/philipp-schmid-a6a2bb196_kyutais-voice-model-moshi-is-here-moshi-activity-7242197402739253248-StiT/?utm_source=share&utm_medium=member_android) 142 | - [Deploying Speech-to-Speech on Hugging Face](https://huggingface.co/blog/s2s_endpoint) 143 | 144 | ### ASR with LLMs: 145 | - [Listening with LLM](https://paul.mou.dev/posts/2023-12-31-listening-with-llm/) 146 | 147 | ### Speech Language Modeling: 148 | - [An Empirical Study of Language Model Integration for Transducer based Speech Recognition](https://arxiv.org/pdf/2203.16776.pdf) 149 | - [AudioPaLM](https://google-research.github.io/seanet/audiopalm/examples/) 150 | 151 | ### Persian ASR Repos: 152 | - [wav2vec2-fa](https://github.com/Hamtech-ai/wav2vec2-fa) 153 | - [Shenasa-ai Speech2Text](https://github.com/shenasa-ai/speech2text) 154 | - [SOTA Persian ASR on Common Voice](https://paperswithcode.com/sota/speech-recognition-on-common-voice-persian) 155 | - [Wav2Vec2 Large-xlsr Persian](https://huggingface.co/m3hrdadfi/wav2vec2-large-xlsr-persian) 156 | - [Wav2Vec2 Large-xlsr Persian (v3)](https://huggingface.co/m3hrdadfi/wav2vec2-large-xlsr-persian-v3) 157 | - [~ 200 Hours Persian ASR Data Set of Shenasa Company](https://github.com/shenasa-ai/speech2text) 158 | - [Persian 380 Hours ASR Data Set](https://www.linkedin.com/posts/amir-pourmand_automatic-speech-recognition-farsi-youtube-activity-7198685117341454337-2sDk?utm_source=share&utm_medium=member_desktop) 159 | 160 | ### Great Resources for Persian ASR Normalization: 161 | - [Wav2Vec2 Large XLSR Persian v3](https://huggingface.co/m3hrdadfi/wav2vec2-large-xlsr-persian-v3) 162 | - [num2fawords: Convert a number into Persian word form](https://github.com/5j9/num2fawords) 163 | - [Parsivar: A Language Processing Toolkit for Persian](https://github.com/ICTRC/Parsivar) 164 | - [num2words](https://github.com/savoirfairelinux/num2words) 165 | 166 | ### Persian based Raw Text Data Sets for LM Training: 167 | - [naab-raw](https://huggingface.co/datasets/SLPL/naab-raw) 168 | 169 | ### `Adapters` Method instead of fine-tuning for Large-Scale ASR models: 170 | - [NeMo Adapters Tutorial](https://github.com/NVIDIA/NeMo/blob/main/tutorials/02_NeMo_Adapters.ipynb) 171 | - [Paper: Parameter-Efficient Transfer Learning for NLP](https://arxiv.org/pdf/1902.00751.pdf) 172 | - [Paper: Efficient Adapter Transfer of Self-Supervised Speech Models for Automatic Speech Recognition](https://arxiv.org/pdf/2202.03218.pdf) 173 | - [Paper: Exploiting Adapters for Cross-lingual Low-resource Speech Recognition](https://arxiv.org/pdf/2105.11905.pdf) 174 | - [Paper: Tiny-Attention Adapter: Contexts Are More Important Than the Number of Parameters](https://arxiv.org/abs/2211.01979) 175 | 176 | ## Diffusion based Methods: 177 | - [Full-band General Audio Synthesis With Score-based Diffusion](https://diffusionaudiosynthesis.github.io/) 178 | 179 | ## Audio Generation: 180 | - [Unconditional Audio Generation Benchmark](https://github.com/gzhu06/Unconditional-Audio-Generation-Benchmark) 181 | - [Realtime Voice Changer](https://github.com/w-okada/voice-changer/tree/master) 182 | 183 | ## Speech Translation: 184 | - [Facebook XLS-R-2B-22-16 Demo](https://huggingface.co/spaces/facebook/XLS-R-2B-22-16) 185 | - [Facebook wav2vec2-xls-r-2b-22-to-16](https://huggingface.co/facebook/wav2vec2-xls-r-2b-22-to-16) 186 | - [Facebook XLS-R-2B-22-16 app code](https://huggingface.co/spaces/facebook/XLS-R-2B-22-16/blob/main/app.py) 187 | 188 | ## G2P (Grapheme2Phoneme): 189 | - [English Grapheme To Phoneme (G2P) Conversion](https://github.com/Kyubyong/g2p) 190 | - [Phonemizer: Simple text to phones converter for multiple languages](https://github.com/bootphon/phonemizer) 191 | - [Epitran: tool for transcribing orthographic text as IPA](https://github.com/dmort27/epitran) 192 | - [PersianG2P](https://github.com/PasaOpasen/PersianG2P) 193 | - [Persian_G2P - link2](https://github.com/AzamRabiee/Persian_G2P) 194 | - [Persian Attention Based G2P](https://github.com/hajix/G2P#attention-based-grapheme-to-phoneme) 195 | - [Tihu Dictionary for Persian Language](https://github.com/tihu-nlp/tihudict) 196 | - [CharsiuG2P: Multilingual G2P in over 100 languages](https://github.com/lingjzhu/CharsiuG2P) 197 | - [Transphone: zero-shot learning based grapheme-to-phoneme model for 8k languages](https://github.com/xinjli/transphone) 198 | 199 | ## Fundamental Notes in Speech Processing & Courses: 200 | - [Deep Learning for Audio (DLA)](https://github.com/markovka17/dla) [**Great Course**] 201 | - [MFCC Tutorial](http://practicalcryptography.com/miscellaneous/machine-learning/guide-mel-frequency-cepstral-coefficients-mfccs/) 202 | - [Sequence Modeling With CTC](https://distill.pub/2017/ctc/) 203 | - [Explanation of Connectionist Temporal Classification](https://sid2697.github.io/Blog_Sid/algorithm/2019/10/19/CTC-Loss.html) 204 | - [D2L Beam Search](https://d2l.ai/chapter_recurrent-modern/beam-search.html) 205 | - [D2L Attention Mechanisms](https://d2l.ai/chapter_attention-mechanisms/index.html) 206 | - [Introduction to Speech Processing](https://wiki.aalto.fi/display/ITSP/Introduction+to+Speech+Processing) [Good] 207 | - [Audio Signal Proessing for Machine Learning](https://github.com/musikalkemist/AudioSignalProcessingForML) 208 | - [Deep Learning For Audio With Python](https://github.com/musikalkemist/DeepLearningForAudioWithPython) 209 | - [Deep learning (audio) application: From design to deployment](https://github.com/musikalkemist/Deep-Learning-Audio-Application-From-Design-to-Deployment) 210 | - [ASR 2022](https://github.com/besacier/ASR2022) 211 | - [Hugging Face Audio course](https://huggingface.co/learn/audio-course/chapter0/introduction) 212 | 213 | ## Great Kaldi Tutorials: 214 | - [Kaldi Install for Dummies](https://www.assemblyai.com/blog/kaldi-install-for-dummies/) 215 | - [Kaldi Speech Recognition for Beginners a Simple Tutorial](https://www.assemblyai.com/blog/kaldi-speech-recognition-for-beginners-a-simple-tutorial/) 216 | - [Tutorial on Kaldi for Brandeis ASR course](https://github.com/keighrim/kaldi-yesno-tutorial) 217 | 218 | ## ASR Error Correction: 219 | - [FastCorrect](https://arxiv.org/pdf/2105.03842.pdf) 220 | 221 | ## Source Separation: 222 | - [Deezer source separation library](https://github.com/deezer/spleeter) [Great] 223 | - [Music Source Separation Challenge](https://github.com/haoheliu/2021-ISMIR-MSS-Challenge-CWS-PResUNet) 224 | 225 | ## Sound & Audio Classification: 226 | - [Soxan: Wav2Vec2 for speech recognition](https://github.com/m3hrdadfi/soxan) 227 | 228 | ## Voice Activity Detection (VAD) & Speech Activity Detection (SAD): 229 | - [Silero VAD](https://github.com/snakers4/silero-vad) 230 | - [Voice Activity Detection: Identifying whether someone is speaking or not](https://maelfabien.github.io/project/Speech_proj/#) [**Great**] 231 | - [py-webrtcvad: Python WebRTC Voice Activity Detector (VAD)](https://github.com/wiseman/py-webrtcvad) [also, it seems that it can segment audio files] 232 | - [Pyannote Audio](https://github.com/pyannote/pyannote-audio) 233 | - [Remsi: Remove silence from video files via ffmpeg](https://github.com/bambax/Remsi) 234 | 235 | ## Audio Segmentation: 236 | - [inaSpeechSegmenter](https://github.com/ina-foss/inaSpeechSegmenter) 237 | 238 | ## Extract & Remove Vocals from Song in Audio Files: 239 | - [Moises: Remove or isolate vocals and instruments in any song](https://moises.ai/) 240 | 241 | ## Audio Summarization: 242 | - [Audio Summarization API](https://github.com/ifrankandrade/api) 243 | 244 | ## Spoken Language Recognition: 245 | - [Spoken Language Recognition based on Kaldi](https://github.com/igorsitdikov/lid_kaldi) 246 | 247 | ## Keyword Spotting & Speech Command Recognition: 248 | - [PyTorch based toolkit for speech command recognition](https://github.com/idiap/sparch) 249 | - [Multilingual Few-Shot Keyword Spotting in PyTorch](https://github.com/alefiury/multilingual_kws_pytorch) 250 | 251 | ## Active Learning in ASR: 252 | - [Active learning in speech recognition](https://alphacephei.com/nsh/2021/07/13/active-learning.html) 253 | 254 | ## Audio Pretraining, Representation Learning, and Self-Supervised Pretraining: 255 | - [CLAP: Contrastive Language-Audio Pretraining](https://github.com/LAION-AI/CLAP) 256 | 257 | ## Audio Augmentation: 258 | - [Audiomentations: Audio Data Augmentation](https://github.com/iver56/audiomentations) 259 | 260 | ## Speech Emotion Recognition: 261 | - [Speech Emotion Recognition via wav2vec2](https://github.com/audeering/w2v2-how-to) 262 | - [Speech Emotions Recognition with Convolutional Neural Networks](https://www.analyticsvidhya.com/blog/2021/07/speech-emotions-recognition-with-convolutional-neural-networks/) 263 | - [ShEMO Data Set](https://github.com/mansourehk/ShEMO) 264 | 265 | ## Annotation Tools: 266 | - [audino: open source audio annotation tool](https://github.com/midas-research/audino) 267 | - [ASR Corpus Creator](https://github.com/egorsmkv/asr-corpus-creator) 268 | - [Speech Editing Toolkit](https://github.com/Zain-Jiang/Speech-Editing-Toolkit) 269 | 270 | ## Audio Compression: 271 | - [EnCodec: High Fidelity Neural Audio Compression](https://github.com/facebookresearch/encodec) 272 | - [AI Powered Audio Compression Technique](https://ai.facebook.com/blog/ai-powered-audio-compression-technique/) 273 | 274 | ## Audio Variational Autoencoder (VAE): 275 | - [AudioVAE: VAE implementation for audio data](https://github.com/nikuson/AudioVAE) 276 | 277 | ## Speaker Anonymization: 278 | - [Speaker Anonymization](https://github.com/DigitalPhonetics/speaker-anonymization) 279 | 280 | ## Some ASR & Speech Datasets: 281 | - [Peoples Speech](https://mlcommons.org/en/peoples-speech/) 282 | - [Multilingual Spoken Words](https://mlcommons.org/en/multilingual-spoken-words/) 283 | - [PodcastMix: A dataset for separating music and speech in podcasts](https://github.com/MTG/Podcastmix) 284 | - [Quran Speech to Text Dataset](https://www.openslr.org/132/) 285 | - [WavCaps: A ChatGPT-Assisted Weakly-Labelled Audio Captioning Dataset](https://github.com/xinhaomei/wavcaps) 286 | - [YODAS dataset: massive youtube speech dataset with 370k hours across 140 languages (about 100TB)](https://huggingface.co/datasets/espnet/yodas) 287 | 288 | ## Voice Conversion: 289 | - [Seed-VC: zero-shot voice conversion](https://github.com/Plachtaa/seed-vc) 290 | 291 | # Interesting Ideas about Startups with ASR: 292 | It is interesting how quickly people implement ideas. Like the one of `podcast transcript` with Whisper. Here is a selection: 293 | - [podscript](https://podscript.ai/) 294 | - [podtext](https://podtext.ai/) 295 | - [podscription](https://podscription.app/) 296 | - [podsearch](https://podsearch.page/) 297 | - [Some Discussion Notes about above links](https://news.ycombinator.com/item?id=34727695) 298 | - [Vapi: Voice AI for any application](https://vapi.ai/) [Great] 299 | 300 | # Other: 301 | - [Neural Target Speech Extraction (TSE)](https://butspeechfit.github.io/tse_tutorial/) 302 | - [Audio Self-supervised Learning: A Survey](https://arxiv.org/abs/2203.01205) 303 | - [AI Audio Startups](https://github.com/csteinmetz1/ai-audio-startups) 304 | - [Facestar: High quality audio-visual recordings of human conversational speech](https://github.com/facebookresearch/facestar) 305 | - [Fast Infinite Waveform Music Generation](https://github.com/marcoppasini/musika) 306 | - [Nvidia Speech AI Summit 2022](https://www.nvidia.com/en-us/events/speech-ai-summit/) 307 | - [Poly AI](https://poly.ai/) [Interesting Company] 308 | - [uberduck: Open Source Voice AI Community](https://uberduck.ai/) 309 | - [How To Build An AI Customer Service Bot](https://www.youtube.com/watch?v=87ZX56RSamA) 310 | - [podcastfy: Open Source API alternative to NotebookLM's podcast](https://github.com/souzatharsis/podcastfy) 311 | --------------------------------------------------------------------------------