Skip to content
View leosky01's full-sized avatar
  • Rome

Block or report leosky01

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
leosky01/README.md

Leonardo Schiavoni

Machine Learning Engineer / Data Scientist — I build ML systems end-to-end: from raw data and feature engineering to trained, calibrated and deployed models. Physics & Astronomy background (Sapienza University of Rome).

Featured projects

Project What it is
ATP-Prediction-Model Tennis match prediction: PyTorch MLP + custom Elo + isotonic calibration. V3 research track: 188 engineered features, LightGBM+NN ensembles, temporal splits, market-disagreement modeling — with honest metrics
chest-xray-classification Multi-label chest X-ray classification (15 pathologies): Swin Transformer V2, asymmetric losses, Optuna HPO, GradCAM interpretability
mri-quality-assessment Automatic MRI quality scoring (6 classes): focal-loss CNNs + Optuna, PyQt5 DICOM annotation GUI, exploratory SimCLR self-supervised track
tyre-lap-time-prediction Regression on synthetic NASCAR data: paper-lite feature-package comparison — linear baselines win, and the report says why
NASCAR-AI-Strategy-Engine Race strategy engine: XGBoost caution model, GAM tire models, Monte Carlo simulation, bootstrap decision analysis, live Streamlit dashboard (140 tests)
AdaptiveFraudAgents Reply Code Challenge 2026: hybrid rules + multimodal LLM agents (GPT-4o investigation, Gemini audio transcription) with cost-aware design and LLM observability

Toolbox

ML/DL PyTorch · PyTorch Lightning · scikit-learn · LightGBM · XGBoost · Optuna · timm Data pandas · NumPy · SQLAlchemy · ETL pipelines Product Streamlit · FastAPI · Flask · Docker Domain medical imaging (X-ray, MRI/DICOM) · sports analytics · tabular prediction

How I work

Methodology over leaderboards: proper temporal validation, leakage checks, probability calibration, class-imbalance handling, interpretability — and results reported as they are, including the modest ones.

Contact

Popular repositories Loading

  1. ATP-Prediction-Model ATP-Prediction-Model Public

    Tennis match prediction: PyTorch MLP + custom Elo + isotonic calibration; V3 research track with 188 features, LightGBM+NN ensembles and temporal splits

    Jupyter Notebook 1

  2. NASCAR-AI-Strategy-Engine NASCAR-AI-Strategy-Engine Public

    NASCAR race strategy decision engine: XGBoost + GAM models, Monte Carlo simulation, bootstrap decision analysis, live Streamlit dashboard

    Python

  3. AdaptiveFraudAgents AdaptiveFraudAgents Public

    Reply Code Challenge 2026: hybrid rules + multimodal LLM fraud agents (GPT-4o, Gemini audio) with cost-aware design and LLM observability

    Python

  4. chest-xray-classification chest-xray-classification Public

    Multi-label chest X-ray classification (15 pathologies) with Swin Transformer V2 — custom asymmetric losses, Optuna HPO, GradCAM interpretability

    Jupyter Notebook

  5. tyre-lap-time-prediction tyre-lap-time-prediction Public

    Supervised regression on synthetic NASCAR race data: paper-lite feature-package comparison (Linear Regression vs Random Forest vs Gradient Boosting), full-laps vs green-laps experiments

    Jupyter Notebook

  6. mri-quality-assessment mri-quality-assessment Public

    Automatic quality scoring of dorsal-spine MRI series (6 classes): supervised CNN/ResNet pipeline with focal loss + Optuna, PyQt5 DICOM annotation GUI, and an exploratory SimCLR self-supervised track

    Python