gpu-training

基于 Tensorflow，仿 Scikit-Learn 设计的深度学习自然语言处理框架。支持 40 余种模型类，涵盖语言模型、文本分类、NER、MRC、知识蒸馏等各个领域

PyTorch implementation of DDPM (Denoising Diffusion Probabilistic Models) for image generation. Includes U-Net with attention, DDIM sampling, EMA training, and CelebA dataset support.

Final project for the UPC Postgrau in Artificial Intelligence & Deep Learning, showcasing a SELF-DRIVING demo built around lane detection and object detection models. Integrates LaneNet (with ENet backbone and BDD100K transforms), a custom Mask R-CNN segmentation pipeline, and Faster R-CNN with ResNet50-FPN.

AI-Farm is a distributed deep learning training framework that enables efficient model training across multiple machines. It provides a scalable infrastructure with real-time monitoring through a web admin panel, adaptive task distribution, and support for both CPU and GPU training.

LoRA fine-tuning pipeline for LLMs with Ollama integration. Supports CPU/GPU training, distributed training with PyTorch DDP, and GGUF conversion.

MetaClaw fork: self-hosted GPU training (GRPO + LoRA + vLLM), per-agent skill isolation, multi-agent mode routing, training engine bug fixes. Train your AI agent on your own A100s. No cloud dependency.

Web safety project that scans online content and helps identify harmful, offensive, or bullying messages directly in the browser.

Production-ready PyTorch framework for distributed deep learning training with Ray & Horovod backends. Optimized for computer vision and time series on Kubernetes clusters.

Data processing, Machine Learning codes and Training scripts on Jean Zay High Performance Computing (HPC) remote GPUs

Let your coding agent run GPU training on spot (preemptible) instances: it launches the run, monitors progress and spend, recovers from preemption and cleans up.

Developed an end-to-end LLM pipeline that extracts Python code from GitHub, builds a high-quality dataset, fine-tunes CodeGen, and performs advanced code generation with DeepSeek. Demonstrates strong capabilities in LLM training, data engineering, and model optimization.

Whatsapp Assistant using FastAPI, Google Gemini, WhisperAI, Twilio and Ngrok

1D CNN for customer complaint text classification using PyTorch.

Public OpenClaw skills for benchmark-gated CV experimentation and SOTA campaigns across Colab, Kaggle, and GPU VMs.

EGM is a deep learning tool that learns your image editing style from raw/edited pairs and applies it to new images. It uses a Pix2Pix (conditional GAN) architecture.

Medical X-ray image classifier fine-tuning ResNet-18 with PyTorch.

A packaged, reproducible training pipeline for [piper1-gpl](https://github.com/OHF-Voice/piper1-gpl) — dataset prep through to a deployable Home Assistant voice. This packages the current one, with every Linux build trap pre-solved, and adds a ROCm/gfx1151 variant that doesn't otherwise exist.

Pollen Studio is a local web app for annotating robotics datasets and training Limelight TFLite detectors. Upload photos, review predictions, and label pollen and blue/red Nectar. Supports NVIDIA GPU training on Linux and Windows via WSL2. Built by Harambe Cartel, Team 20912.

Evidence-gated quantitative forecasting with source-age-aware models, GPU experiments, PIT provenance, calibration, and model-risk diagnostics.

Autonomous AI research agents for single-GPU nanochat training — automated experiment design, execution, and analysis