Awesome GitHub RepositoriesTraining Optimizations

This project is a fine-tuning framework and training pipeline designed to optimize and adapt large language and vision models. It provides a specialized toolkit for parameter-efficient tuning and supervised learning, serving as both a trainer for multimodal models and a deployment tool for serving fine-tuned models via high-performance inference engines. The framework focuses on reducing memory and compute requirements by updating a small subset of model parameters. It supports a wide range of adaptation strategies, including vision-language model training to align text, image, video, and aud

DeepSpeed is a high-performance library designed to scale deep learning model training and inference across massive clusters of GPUs and compute nodes. It provides a comprehensive suite of tools for distributed training, enabling the execution of models that exceed the memory capacity of single devices through advanced parameter partitioning, pipeline-based model parallelism, and memory-efficient state offloading. The framework distinguishes itself through specialized communication-efficient optimizers and hardware-aware acceleration techniques. By utilizing gradient compression, quantization

ControlNet is a framework for structural image generation that extends pre-trained diffusion models with neural network architectures designed for precise spatial control. By injecting structural guidance directly into the latent-space denoising process, the system enables users to enforce geometric or semantic constraints on generated outputs while maintaining style consistency. The framework distinguishes itself through a weight-locked copying mechanism that preserves the integrity of the original model while introducing new control signals. It supports multi-condition synthesis, allowing f

This project is an open-source educational resource providing structured, step-by-step guides for fine-tuning large language models. It focuses on adapting pre-trained transformer-based causal models to custom datasets, enabling users to transfer specific writing styles or domain knowledge into generative AI models. The repository distinguishes itself by emphasizing parameter-efficient training techniques, specifically low-rank adaptation. By providing practical implementations for updating only a small subset of model weights, it allows for the customization of massive neural networks on con

Swin-Transformer is a deep learning framework designed for training and deploying hierarchical vision transformer models. It serves as a research library and toolkit for computer vision tasks, providing the infrastructure to build models that replace standard convolution operations with sliding window self-attention mechanisms. By utilizing a multi-scale feature hierarchy, the framework enables the processing of visual data at varying resolutions and spatial scales. The project distinguishes itself through its implementation of shifted window partitioning, which facilitates global information

PaddleDetection is an object detection framework designed for the end-to-end development, training, and deployment of computer vision models. It provides a comprehensive library of modular neural network architectures and pipelines that support object detection, instance segmentation, and multi-object tracking tasks. The project distinguishes itself through a configuration-driven approach that decouples model components like backbones and heads, allowing for the flexible assembly of custom vision workflows. It incorporates advanced techniques such as anchor-free detection logic, joint detecti

YOLOv7 is a PyTorch vision library and real-time inference engine designed for object detection, human pose estimation, and instance segmentation. It provides a framework for detecting and locating multiple objects within images or video streams using neural networks. The system includes tools for custom model training and fine-tuning, allowing pre-trained weights to be adapted to specialized datasets via transfer learning. It also supports model weight export and format conversion to facilitate deployment on production servers and embedded edge devices.

kohya_ss is a graphical user interface and workbench for fine-tuning diffusion models, specifically designed for Stable Diffusion. It provides a suite of tools for training generative AI models, including specialized interfaces for creating Low-Rank Adaptation weights and training ControlNet spatial control networks. The project distinguishes itself through integrated VRAM usage optimization and hardware acceleration, featuring specific support for Intel GPUs via XPU-accelerated libraries. It implements parameter-efficient training methods and memory-saving techniques like gradient checkpoint

AutoGluon is an automated machine learning framework and multimodal library designed to automate the end-to-end pipeline from data preprocessing to high-accuracy model training and validation. It functions as an automated model trainer for tabular, image, text, and time series data, as well as a tool for time series forecasting and foundation model finetuning. The project is distinguished by its ability to jointly process and fuse different data types, allowing for the construction of multimodal neural networks that integrate images, text, and structured tables. It supports zero-shot inferenc

SlowFast is a PyTorch video understanding framework and spatiotemporal neural network library. It serves as a toolset for video action recognition, enabling the training and evaluation of models designed to classify complex activities and objects within video sequences. The framework is distinguished by its use of dual-pathway spatiotemporal sampling to capture both slow and fast motions. It supports self-supervised video learning for pre-training models on unlabeled data and employs multigrid spatiotemporal training to optimize learning across multiple spatial and temporal resolutions. The

This project is a neural machine translation system used to build models that automatically translate text from one language to another. It utilizes sequence-to-sequence modeling to transform variable-length input sequences into corresponding output sequences. The system implements bidirectional recurrent neural network encoding and attention mechanisms to capture contextual information and focus on specific parts of the source text during translation. To manage training and inference, it employs separate computational graphs and supports distributing model layers across multiple GPU devices.

Liger-Kernel is a collection of pre-built fused Triton kernels and patching utilities designed to accelerate large language model training. It provides drop-in kernel replacements for common LLM operations such as RMSNorm, cross-entropy loss, and attention, enabling increased throughput and reduced memory usage while preserving bitwise-exact gradients. The project serves as a toolkit for composing custom model architectures from individual optimized kernels and for patching pre-existing models with minimal code changes. The project distinguishes itself through its ability to perform runtime m

Neural Prophet is a PyTorch-based time series forecasting library designed for interpretable machine learning. It serves as a decomposition framework that breaks signals into constituent parts such as autoregressive effects, piecewise linear trends, and Fourier-based seasonality to predict future values. The project distinguishes itself by combining neural networks with traditional algorithms to produce forecasts that explain underlying trend drivers. It features a global time series modeling approach, allowing a single model to be trained across multiple simultaneous series to share learned

Acest proiect este un toolkit de deep learning conceput pentru separarea surselor audio și recuperarea informațiilor muzicale. Acesta oferă un framework pentru descompunerea semnalelor audio polifonice în componente distincte, cum ar fi vocea, tobele și basul, prin procesarea formelor de undă brute prin arhitecturi de rețele neuronale. Biblioteca permite utilizatorilor să antreneze modele de separare personalizate sau să le ajusteze pe cele existente pentru a îmbunătăți acuratețea pe seturi de date audio specifice. Suportă întregul ciclu de viață al modelului, inclusiv convertirea audio-ului brut în formate structurate, indexate, pentru a optimiza încărcarea datelor și eficiența antrenării. Sistemul include capabilități pentru extragerea caracteristicilor în domeniul frecvenței și pipeline-uri de date bazate pe tensori pentru a facilita sarcini complexe de procesare a semnalelor. Este structurat pentru a susține fluxuri de lucru orientate spre cercetare, permițând rafinarea iterativă a modelelor prin persistența bazată pe checkpoint-uri și orchestrarea automatizată a antrenării.