Skip to content
View Alireza-Motazedian's full-sized avatar
🙂
smile
🙂
smile

Organizations

@UNH-Machine-Learning-in-Space-Physics @Homaei-Research-Group @Second-Harmonic-Generation

Block or report Alireza-Motazedian

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
Alireza-Motazedian/README.md

Alireza Motazedian     Linkedin Google Scholar ORCID

🚀 🔭 Scientific Machine Learning | Space Physics | Space Weather Modeling

🔬 Scientific Machine Learning → Physics‑Informed Models

  • Inverse problems & Tikhonov regularization
  • Ensemble & time‑series models with uncertainty quantification
  • Physics-constrained reconstruction of geophysical fields

🌌 Space Physics & Space Weather → Model the Geospace

  • Real-time IMEF & Cross Polar Cap Potential (CPCP) prediction
  • Electric-field reconstruction via field-line mapping & inversion
  • Geomagnetically induced current (GIC) modeling

⏱ Time Series & Forecasting → Nonstationary Systems

  • LSTM, TCN, ARIMA & Prophet models
  • Feature engineering & noise-robust preprocessing
  • Storm-time forecasting of geophysical dynamics

🛠 ML Engineering → From Research to Deployment

  • End-to-end ML pipelines & experiment tracking (MLflow)
  • Automated data ingestion & scheduled inference
  • Deployment via Docker, systemd & GitHub Actions
  • GPU acceleration & performance benchmarking

⚙ Physics & Computational Modeling

  • FORTRAN/Python simulators & finite-difference methods
  • PDE-based modeling of electromagnetic systems
  • Nonlinear optics & second-harmonic generation
  • High-performance scientific computing

🎓 Teaching & Mentorship

  • Instructor/TA at UNH Physics Department
  • Guided problem-solving & structured learning design
  • Mentorship in numerical modeling & ML

Skills

Skills Technologies
ML Engineering Docker MLflow
Deep Learning TensorFlow PyTorch
Machine Learning scikit-learn
Cloud Platforms AWS Azure Google Cloud
Time Series Analysis   TCN Prophet LSTM Statsmodels ARIMA SARIMA
Pkg / Env Managers Conda Venv virtualenv pip
Python Python Pandas NumPy Statsmodels ipywidgets fastparquet
Git Git GitHub GitLab
Visualization Draw.io Visio Matplotlib Seaborn ggplot2
Editors Cursor VS Code Google Colab JupyterLab Jupyter Notebook PyCharm

Pinned Loading

  1. Classification_API_with_PostgrSQL Classification_API_with_PostgrSQL Public

    End-to-End Classification APIs — A curated suite of production-ready classification models, each deployed via FastAPI with PostgreSQL and Docker integration.

    Jupyter Notebook 1

  2. Dockerized_ANN Dockerized_ANN Public

    End-to-end Artificial Neural Networks: built from scratch and with modern ML libraries. Includes custom training loops, real-world datasets, model tuning, and insightful visualizations.

    Jupyter Notebook 1

  3. Regression_API_with_PostgrSQL Regression_API_with_PostgrSQL Public

    Modular Regression Models — A curated suite of standalone repositories showcasing linear, polynomial, and advanced regression techniques, each built with clean, production-ready pipelines and real-…

    Jupyter Notebook 1

  4. Dockerized_CNN Dockerized_CNN Public

    Curated Convolutional Neural Networks — A focused collection of CNN projects covering image classification, feature extraction, and model optimization with clean, modular, and scalable implementat…

    Jupyter Notebook 1

  5. Nonlinear_Optics_Second_Harmonic_Generation Nonlinear_Optics_Second_Harmonic_Generation Public

    Computational SHG Modeling — A complete Fortran-based framework for simulating thermal effects in Second Harmonic Generation using FDM and coupled physics.

    Gnuplot 2

  6. Dockerized_LSTM Dockerized_LSTM Public

    Mastering LSTM Networks — A focused collection of high-impact projects for sequence modeling, time series forecasting, and pattern recognition using robust, well-documented, and production-grade code.

    Jupyter Notebook 1