Skip to content
View mehdighelich1379's full-sized avatar
  • iran

Block or report mehdighelich1379

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, 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
mehdighelich1379/README.md

Typing SVG

Data Scientist | Deep Learning Enthusiast | Electronics Engineer

Late night coding in dark room

Follow Mehdi on GitHub

🛠️ Skills & Tools

🧠 Machine Learning & Deep Learning Skills

  • Supervised & Unsupervised Learning
  • Model Evaluation: Accuracy, Precision, Recall, AUC
  • Data Balancing: SMOTE, Class Weights
  • Dimensionality Reduction: PCA
  • Outlier Detection: LOF
  • Pipeline Design & Feature Engineering

📌 Algorithms & Techniques

  • Classification: SVM, Logistic Regression, Decision Tree, Random Forest, CatBoost, XGBoost, LightGBM
  • Clustering: KMeans, DBSCAN
  • Deep Learning: CNN (used in medical image classification projects such as chest X-rays, kidney stones, and multi-class eye diseases)

🧬 Currently Learning

  • Deep Learning (more in-depth: architectures, optimization, and regularization)
  • Sequence Models: LSTM, GRU, SimpleRNN
  • Transformers & Attention Mechanisms
  • Deployment of AI Projects: End-to-end workflows including model packaging, Streamlit apps, Docker, and cloud deployment (e.g. Heroku, AWS)

📈 GitHub Stats

🌐 Connect with Me

🚀 Learning daily, building constantly.

Pinned Loading

  1. Smart-Fraud-Detection Smart-Fraud-Detection Public

    A complete fraud detection pipeline using ML models (CatBoost, XGBoost, LightGBM), class weighting, SMOTE, and custom feature engineering. Achieved strong recall and precision balance for real-worl…

    Jupyter Notebook

  2. churn-prediction-ml-dl churn-prediction-ml-dl Public

    A complete machine learning and deep learning pipeline for customer churn prediction. Includes preprocessing, model training (RandomForest, XGBoost, CatBoost, Keras), evaluation (confusion matrix, …

    Jupyter Notebook

  3. Insurance-Loss-Prediction Insurance-Loss-Prediction Public

    This project predicts credit card fraud using machine learning techniques. The dataset contains transaction records with features like amount, merchant, and user behavior. The goal is to classify t…

    Jupyter Notebook

  4. Heart_Disease Heart_Disease Public

    This project focuses on predicting the likelihood of heart disease using machine learning techniques. The dataset includes medical features like age, blood pressure, cholesterol, and heart rate. Th…

    Jupyter Notebook

  5. OCT-Retinal-Disease-Detection-CNN OCT-Retinal-Disease-Detection-CNN Public

    Deep learning project for retinal disease classification using OCT images. Leveraging MobileNetV2 and transfer learning, the model classifies CNV, DME, DRUSEN, and NORMAL with 93% accuracy. Include…

    Jupyter Notebook

  6. Chest-Xray-Pneumonia-Prediction Chest-Xray-Pneumonia-Prediction Public

    uses a Convolutional Neural Network (CNN) to classify chest X-ray images into two categories: Pneumonia and Normal. The model is built from scratch using layers like Conv2D, MaxPool2D, and Dropout …

    Jupyter Notebook