A comprehensive collection of machine learning and data science projects developed during my professional internship
This repository showcases real-world data engineering and machine learning solutions, demonstrating expertise in classification, regression, computer vision, and predictive analytics.
| # | Project | Domain | Key Techniques | Status |
|---|---|---|---|---|
| 1 | Email Spam Classification | NLP | Naive Bayes, TF-IDF, Feature Engineering | ✅ Complete |
| 2 | House Price Prediction | Regression | Linear Regression, Feature Selection | ✅ Complete |
| 3 | Image Classification CIFAR | Computer Vision | CNN, Deep Learning, Image Processing | ✅ Complete |
| 4 | Image Processing Task 5 | Computer Vision | OpenCV, Filtering, Transformations | ✅ Complete |
| 5 | Customer Churn Prediction | Business Analytics | Classification, Logistic Regression, EDA | ✅ Complete |
- Machine Learning: Scikit-learn, TensorFlow, Keras
- Data Processing: Pandas, NumPy
- Visualization: Matplotlib, Seaborn
- Computer Vision: OpenCV, PIL
- NLP: NLTK, TF-IDF Vectorizer
python >= 3.8
pip >= 21.0-
Clone the repository
git clone https://github.com/SajjadKhanYousafzai/DEP.git cd DEP -
Create a virtual environment (recommended)
python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate
-
Install dependencies
pip install numpy pandas scikit-learn tensorflow opencv-python matplotlib seaborn jupyter nltk
-
Navigate to any project folder
cd "Email Spam Classification" jupyter notebook
Objective: Build an intelligent email filter to classify messages as spam or legitimate.
Features:
- Text preprocessing and cleaning
- TF-IDF feature extraction
- Multiple classifier comparison (Naive Bayes, SVM, Random Forest)
- Model performance evaluation
Key Metrics: Accuracy, Precision, Recall, F1-Score
Objective: Develop a regression model to predict house prices based on various features.
Features:
- Exploratory Data Analysis (EDA)
- Feature engineering and selection
- Linear and polynomial regression
- Cross-validation and hyperparameter tuning
Key Techniques: Regression Analysis, Feature Scaling, Model Evaluation
Objective: Create a deep learning model to classify images from the CIFAR dataset.
Features:
- Convolutional Neural Network (CNN) architecture
- Image augmentation techniques
- Transfer learning implementation
- Model optimization and fine-tuning
Dataset: CIFAR-10/CIFAR-100
Objective: Apply advanced image processing techniques for enhancement and analysis.
Features:
- Edge detection and contour analysis
- Image filtering and noise reduction
- Morphological operations
- Color space transformations
Tools: OpenCV, PIL, NumPy
Objective: Predict customer churn to enable proactive retention strategies.
Features:
- Customer behavior analysis
- Feature importance identification
- Classification model development
- Business insights generation
Impact: Helps businesses reduce customer attrition
Each project folder contains:
- ✅ Complete source code with detailed comments
- 📓 Jupyter notebooks with step-by-step analysis
- 📈 Visualization of results and model performance
- 📄 Documentation and findings
Through these projects, I gained hands-on experience in:
- ✔️ End-to-end ML pipeline development (data collection → deployment)
- ✔️ Feature engineering and selection for optimal model performance
- ✔️ Model evaluation using appropriate metrics and validation techniques
- ✔️ Deep learning architectures for computer vision tasks
- ✔️ Real-world problem solving with data-driven approaches
- Deploy models as REST APIs using Flask/FastAPI
- Create interactive dashboards with Streamlit
- Implement MLOps practices (model versioning, monitoring)
- Add more advanced deep learning projects
- Containerize applications with Docker
Sajjad Khan Yousafzai
📧 Feel free to reach out for collaborations or discussions on machine learning and data science!
If you find these projects helpful or interesting, please consider giving this repository a star! ⭐
Made with ❤️ during my Machine Learning Internship
Last Updated: January 2026