Showing 30 of 1,459 · Category: Machine Learning
Credit scorecard (0.60 Gini) with a cost-based cut-off and an approve/decline tool with plain-language reasons
Machine learning toolkit for battery cathode materials: predicts formation energy, band gap and average voltage from chemical formula (RF+HGB ensemble, 5-fold CV, data->features->train->screen->predict).
Household electricity forecasting and CO₂ estimation using machine learning, FastAPI, and React.
Explainable customer churn prediction for e-commerce using XGBoost, SHAP and Streamlit
Quant ML research platform: deep learning on CME Level-3 order-book data (CNN-Mamba, PatchTST), Rust feature pipeline, Ray cluster + MLflow, walk-forward/OOT validation, RL execution, options & systematic strategy research
End-to-end heart disease classification project using Logistic Regression, KNN and Random Forest with hyperparameter tuning and model evaluation.
知衡：AI Agent 驱动的算法工程工作空间，集成知识图谱、代码生成、自修复、独立验证与能力版本管理。 AxiomForge: Evidence-driven agent engineering for algorithm generation, validation, repair and reusable knowledge.
Live silent-word lip reading in the browser: MediaPipe face landmarks + a small model trained on the GRID corpus, with teach-your-own words. Runs fully on-device.
This is an RL training group to find a policy for an n-link cartpole.
Regression tests for images sent to vision LLM APIs: pixel-identical vs lossy input variants, variance-aware scoring, CI gates.
PRISM: Agentic Enterprise Governance Tower for Procurement Risk — Built for ET AI Hackathon 2026 x Accenture. Features 5-agent pipeline, ML risk scoring, TF-IDF RAG policy retriever, fail-closed human gate, and SHA-256 audit chaining.
training materials
Indonesian receipts and invoices to validated Excel: OCR, classification, rule/LLM extraction, calibrated confidence, human review. Portfolio AI/ML project.
金融データ分析ポートフォリオ｜SQL・Python・Power BI
Predicting which Philippine municipalities stay without power after typhoons, using NASA Black Marble nighttime lights, Google Earth Engine and machine learning, with spatially explicit tests of transfer to unseen storms.
End-to-end Data Engineering & ML Pipeline with PySpark (Medallion Architecture) on Madrid meteorological data.
Non-invasive haemoglobin screening from smartphone conjunctiva images, with a skin-tone fairness audit. Browser-side inference.
Gait speed estimation in older adults from wearable IMU data with a GRNN (MATLAB). Published at IWAIPR 2021
Reproducible code for our IEEE ICISS 2026 paper comparing SMOTE and SMOTE-ENN for stroke prediction (logistic regression, SVM, XGBoost). Table IV reproduces the original notebook exactly on Linux CI; cross-validation, SHAP and calibration are rebuilt. Shows that resampling mostly shifts recall, not ranking quality.
Top Distributed AI On Kubernetes (Opensource) Star if you like it!
API and Deep Learning application for froth flotation bubble segmentation, covering dataset preparation, model training, inference, and API serving with FastAPI.
On Time is an Alexa+ add-on built as an MCP server (spec 2025-11-25) that keeps your lights in step with the sun, the seasons, and you, easing through Daylight Saving Time.
Materials quality copilot on Snowflake Cortex: data pipeline, Cortex Search, Cortex LLM, Snowflake ML, Node.js API and React UI over 4Cr13 axle / SEM data.
Interactive 3D solar-system simulator combining deterministic Light physics with an evaluated Machine-Learning surrogate.
Chronological-validation forecasting of US inflation & unemployment (FRED): ARIMA/SARIMA vs ML against honest baselines, with Diebold–Mariano testing
Fine-tune, merge, and run a local DevOps AI assistant using SmolLM2 1.7B, LoRA, Hugging Face Transformers, and Apple MPS.
Top Machine Learning Explainability XAI (Opensource) Star if you like it!
Predicting 12-hour soil-moisture irrigation risk from field sensor, weather and crop data (Flanders, 2021-2023). Leakage-safe ML pipeline with explainability, calibration and robustness tests.
AI-based crop recommendation website for Indian farmers. Flask, SQLite and a Random Forest model suggest the best crops for your soil and weather, with growing guides for 70 crops.