Aprendizaje automático · GitHub

SciML-Bench Benchmarks for Scientific Machine Learning (SciML), Physics-Informed Machine Learning (PIML), and Scientific AI Performance

High Performance Computation for N-D Tensors in .NET, identical API to NumPy.

• NumPy-style NDArray - N-dimensional arrays with shape, strides, offsets, slicing, and view semantics. Start with NDArray fundamentals and NDArray.
• Broadcasting - NumPy-style shape expansion without materializing repeated values. See Broadcasting.

Stock Price Prediction using Machine Learning Techniques

VintageVoice — open-source TTS for historical speech patterns: Transatlantic accent, newsreel narrators, Edison-cylinder-era voices (1880s-1960s). Machine learning as cultural heritage preservation.

Free, open-source (AGPL) mosaic restoration for Windows: detects pixelated regions in video and restores them with a temporal model, plus whole-frame upscaling. NVIDIA RTX edition (TensorRT, NVDEC/NVENC, RTX Super-Res). AMD and Intel Arc editions in sibling repos.

A curated list of awesome resources related to capsule networks

PyTorch 기반 한국어 LLM 구현 및 학습 프로젝트

Multi-class confusion matrix library in Python

Single-Shot Refinement Neural Network for Object Detection, CVPR, 2018

Open-Source AI Camera Skills Platform, AI NVR & CCTV Surveillance. Local VLM video analysis with Qwen, DeepSeek, SmolVLM, LLaVA, YOLO26. LLM-powered agentic security camera agent — watches, understands, remembers & guards your home via Telegram, Discord or Slack. Pluggable AI skills. OpenAI, Google, Anthropic or local AI. Runs on Mac Mini & AI PC.

Easy-to-use,Modular and Extendible package of deep-learning based CTR models .

Fine-tuned Qwen2-VL-7B for LaTeX OCR using LoRA and Unsloth on the LaTeX OCR dataset. Built augmentation pipeline (rotation, noise, contrast jitter), ran LoRA rank sweep (r=8/16/32), and evaluated across CER, Token F1, BLEU-4, and Exact Match. Deployed as a Gradio Space with live metric computation.

Open-Unmix - Music Source Separation for PyTorch

FSQ K-line foundation model for China A-share stock selection, with a containerised multi-factor research platform: data governance, factor research and observability.

• Finance-native pre-training. 100 % financial corpus: 3.06 billion raw K-lines across 7 frequencies (1-minute to weekly), including 5,872 delisted stocks to avoid survivorship bias.
• FSQ parallel tokenization. Four bounded scalars with levels [8, 8, 8, 5] give 2,560 implicit states with no codebook, so codebook collapse cannot happen. Limit-up/down bars, gaps and volume spikes map to edge levels instead of being flattened. 71.4 % of joint states are used, with normalised entropy 0.87.