🧠 AI/ML Tutorial Projects Repository

Your Complete Journey from Zero to AI Hero 🚀

A comprehensive, hands-on collection of AI & Machine Learning tutorials covering everything from fundamentals to cutting-edge techniques

Quick Start • Topics Covered • Project Structure • Contribute

🎯 About This Repository

Welcome to the ultimate AI learning repository! This carefully crafted collection takes you on a complete journey through the fascinating world of Artificial Intelligence and Machine Learning. Whether you're a complete beginner or looking to master advanced techniques, this repository has you covered.

💡 What Makes This Special?

🎓 Structured Learning Path: Follow a progressive curriculum from basics to advanced topics
💻 Hands-On Projects: Real, working code you can run and modify
🌐 Google Colab Ready: Most notebooks include Colab links for GPU-powered training
🇹🇷 Turkish & English: Bilingual support for better understanding
📊 Real Datasets: Work with actual data including images, text, and time series
🎨 Visual Learning: Rich visualizations and examples throughout

⚡ Pro Tip: For computationally intensive projects, use Google Colab with GPU enabled. Don't forget to save your trained models locally before your session expires!

📚 Topics Covered

🔰 Foundations & Prerequisites

Click to expand core fundamentals

1-Prerequisites

Master the essential Python libraries for data science:

📊 NumPy Tutorial: Mathematical operations and array manipulation
🐼 Pandas Tutorial: Data manipulation and analysis with real datasets (Northwind, JSON)
📈 Matplotlib: Data visualization fundamentals

2-MachineLearningBasics

Statistical Foundations (Temelİstatistik.ipynb)
Regression Analysis (Regresyon.ipynb)
Practical examples with car datasets

3-YapayZekaTeorikBilgi (AI Theoretical Knowledge)

🧪 Lesson 1: What is AI? History and fundamentals (Perceptron, Neural Networks)
🧪 Lesson 2: Deep dive into ML concepts (Supervised, Unsupervised, Reinforcement Learning)
Rich visualizations of AI/ML/DL relationships

🎯 Classification & Machine Learning

Click to expand ML & classification projects

4-MathValClass (Numerical Classification)

Learn classical ML algorithms for numerical data:

✅ Logistic Regression (LR) - Binary classification with sigmoid functions
✅ Linear Discriminant Analysis (LDA) - Feature combination for separation
✅ K-Nearest Neighbors (KNN) - Simple yet powerful neighbor-based classification
✅ Classification & Regression Trees (CART) - Decision tree methods
✅ Gaussian Naive Bayes (NB) - Probabilistic classification with Bayes theorem
✅ Support Vector Machines (SVM) - Hyperplane-based classification

Projects:

🌸 Iris Classification - The classic ML introduction
📊 Complete classification comparisons

🖼️ Computer Vision & Deep Learning

Click to expand CV projects

5-ImageClass (Image Classification)

Master CNNs and modern architectures:

🚗✈️ Car vs Plane Classifier: Binary image classification with CNNs
🐱🐰🐿️ Cat-Rabbit-Squirrel: Multi-class classification with ResNet
🐱🐶 Cat vs Dog: Classic computer vision challenge
😊😢😡 Facial Expression Recognition (FER): Emotion detection from faces (FER2013 dataset)
🎯 YOLOv11 Classification: Modern classification with YOLO

🏗️ ArchitecturesAdv - Complete Model Zoo Comparison ⭐

A comprehensive comparison of 6 cutting-edge architectures on FER2013 dataset:

🧱 Basic CNN: Classic convolutional architecture baseline
🔗 ResNet: Residual connections for deeper networks
🌳 DenseNet121: Dense connectivity patterns
⚡ EfficientNetB0: Compound scaling for efficiency
👁️ CNN + SE Attention: Squeeze-and-Excitation blocks for channel attention
🔮 Vision Transformer (ViT): Transformer architecture for images

Features:

Side-by-side performance comparison
FLOPs and parameter count analysis
Training time benchmarks
Visual architecture diagrams (visualkeras + plot_model)
Validation accuracy plots across all models
Random prediction comparisons

6-ObjectDetection (Object Detection)

Go beyond classification - locate objects in images:

🎯 YOLOv7 + RoboFlow: Industry-standard object detection
🎯 YOLOv11 Training: Latest YOLO architecture
📦 Custom Dataset Training: Train on your own data with Kaggle datasets

🤖 Advanced Deep Learning

Click to expand advanced topics

7-LSTM (Long Short-Term Memory Networks)

Time series and sequential data:

📈 Univariate LSTM: Flight data prediction
🔮 Sequential pattern learning

8-Generative (Generative AI)

Create new content with AI:

🎨 DCGAN: Deep Convolutional Generative Adversarial Networks
👗 Fashion MNIST generation
🎭 Generator vs Discriminator training dynamics
Multiple architecture variations (28x28, custom sizes)

9-NeuralStyleTransfer

🎨 Transfer artistic styles between images
Neural networks as artists

10-Pix2Pix (Image-to-Image Translation)

Revolutionary paired image translation:

🎨 Image Colorization AI: Turn grayscale to color automatically
🔄 Complete Pix2Pix implementation
🖼️ Paired image dataset processing

📝 Natural Language Processing

Click to expand NLP projects

11-TextClass (Text Classification)

📚 RNN Text Classification: Sentiment and text categorization using Recurrent Neural Networks

13-RAG (Retrieval Augmented Generation)

Modern AI chatbot techniques:

🤖 Ollama on Colab: Run local LLMs in the cloud
📖 RAG implementation for document Q&A
PDF processing and knowledge retrieval

14-WordEmbeddings

The foundation of modern NLP:

🔤 Text → Embedding → Cosine Similarity
🧠 Semantic search implementation
💾 Memory systems for RAG applications
Integration with sentence-transformers

📊 Additional Topics

Click to expand extra content

12-VideoSources (Data Analysis & Visualization)

🌊 CalCOFI Ocean Data: Marine ecosystem analysis
📊 Seaborn Visualizations: Advanced plotting techniques
💰 Tips Dataset: Regression and correlation analysis
🎥 Hand Controls: Computer vision control systems

999-BonusBilgiler (Bonus Knowledge)

🧪 A/B Testing: Statistical hypothesis testing
🎲 German Tank Problem: Bayesian inference example
💼 Interview Q&A: ML interview preparation (mulakat_soru_cevapları.py)
🐍 Python Recap: Complete Python refresher (python_recap.py)
📚 PyCharm cheat sheets and environment setup

DatabaseGen

🗄️ ImageNet dataset management
🔄 Data preprocessing utilities
📦 Archive management tools

🗂️ Project Structure

MlAiTutorialProjects/
│
├── 1-Prerequisites/              # NumPy, Pandas, Matplotlib fundamentals
├── 2-MachineLearningBasics/      # Statistics, Regression basics
├── 3-YapayZekaTeorikBilgi/       # AI theory and concepts (Turkish)
├── 4-MathValClass/               # Classical ML algorithms
├── 5-ImageClass/                 # CNN & image classification
├── 6-ObjectDetection/            # YOLO models & object detection
├── 7-LSTM/                       # Time series with LSTM
├── 8-Generative/                 # GANs & generative models
├── 9-NeuralStyleTransfer/        # Style transfer techniques
├── 10-Pix2Pix/                   # Image-to-image translation
├── 11-TextClass/                 # NLP & text classification
├── 12-VideoSources/              # Data analysis & visualization
├── 13-RAG/                       # Retrieval Augmented Generation
├── 14-WordEmbeddings/            # Text embeddings & semantic search
└── 999-BonusBilgiler/            # Bonus content & interview prep

🚀 Quick Start

1️⃣ Clone the Repository

git clone https://github.com/onuralpArsln/MlAiTutorialProjects.git
cd MlAiTutorialProjects

2️⃣ Choose Your Path

🌟 For Beginners:

Start here → 1-Prerequisites → 2-MachineLearningBasics → 3-YapayZekaTeorikBilgi → 4-MathValClass

🔥 For Computer Vision Enthusiasts:

5-ImageClass → 5-ImageClass/ArchitecturesAdv (⭐ Model Zoo) → 6-ObjectDetection → 9-NeuralStyleTransfer → 10-Pix2Pix

📝 For NLP Lovers:

11-TextClass → 14-WordEmbeddings → 13-RAG

🎨 For Generative AI Fans:

8-Generative → 10-Pix2Pix → 13-RAG

3️⃣ Run with Google Colab (Recommended)

Most notebooks include a "Open in Colab" badge at the top. Click it to run with free GPU!

# Notebooks are ready to run - just execute cells sequentially
# Colab provides free GPU/TPU access for training

4️⃣ Run Locally

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies (varies by project)
pip install numpy pandas matplotlib tensorflow pytorch scikit-learn

# Launch Jupyter
jupyter notebook

💻 Technologies & Frameworks

Category	Technologies
Languages	Python 3.8+
Deep Learning	TensorFlow, Keras, PyTorch
Classical ML	scikit-learn, XGBoost
Data Science	NumPy, Pandas, Matplotlib, Seaborn
Computer Vision	OpenCV, PIL, YOLOv7, YOLOv11
NLP	sentence-transformers, Ollama
Development	Jupyter Notebooks, Google Colab

🎓 Learning Path Recommendations

🌱 Complete Beginner (0-3 months)

Prerequisites (NumPy, Pandas, Matplotlib)
Machine Learning Basics
AI Theoretical Knowledge
Math Value Classification (start with Logistic Regression)
Simple Image Classification (Car vs Plane)

🌿 Intermediate (3-6 months)

Advanced Image Classification (Multi-class, ResNet)
Object Detection (YOLO)
LSTM for time series
Text Classification with RNN
Basic GANs

🌳 Advanced (6+ months)

Architecture Model Zoo (CNN, ResNet, DenseNet, EfficientNet, ViT comparison) ⭐
Pix2Pix & Image Translation
Neural Style Transfer
RAG & Modern NLP
Custom model architecture design & benchmarking

📝 Key Features of Each Section

Section	Difficulty	Time Estimate	Key Takeaway
Prerequisites	⭐ Beginner	1-2 weeks	Master Python data science libraries
ML Basics	⭐⭐ Beginner	2-3 weeks	Understand regression & statistics
Theory	⭐⭐ Intermediate	1 week	Learn AI fundamentals deeply
Classification	⭐⭐ Intermediate	2-3 weeks	Master classical ML algorithms
Image Class	⭐⭐⭐ Intermediate	3-4 weeks	Build CNNs from scratch
Architectures Adv	⭐⭐⭐⭐⭐ Expert	2-3 weeks	Compare 6 architectures scientifically
Object Detection	⭐⭐⭐⭐ Advanced	2-3 weeks	Deploy YOLO models
LSTM	⭐⭐⭐ Intermediate	2 weeks	Predict time series data
GANs	⭐⭐⭐⭐ Advanced	3-4 weeks	Generate synthetic images
Pix2Pix	⭐⭐⭐⭐⭐ Expert	3-4 weeks	Master image translation
RAG	⭐⭐⭐⭐ Advanced	2-3 weeks	Build modern AI chatbots
Text/NLP	⭐⭐⭐ Intermediate	2-3 weeks	Process and classify text
Embeddings	⭐⭐⭐⭐ Advanced	1-2 weeks	Understand semantic search

🔥 Highlighted Projects

🏆 NEW: Complete Architecture Model Zoo ⭐⭐⭐

The ultimate deep learning architecture comparison guide!

📁 Location: 5-ImageClass/ArchitecturesAdv/architec.ipynb
🎯 Architectures: 6 models (CNN, ResNet, DenseNet, EfficientNet, CNN+SE, ViT)
🔬 Metrics: Parameters, FLOPs, training time, validation accuracy
💾 Dataset: FER2013 (28,709 train + 7,178 test images)
📊 Visualizations: Architecture diagrams, performance plots, prediction comparisons
🎓 Perfect for: Understanding trade-offs between model complexity and performance

What makes it special:

🧪 Scientific comparison with consistent methodology
📊 Quantitative metrics (not just accuracy!)
🎨 Beautiful architecture visualizations using visualkeras
🔍 Includes modern architectures (Vision Transformers!)
⚡ Performance profiling with FLOPs calculation
🎯 Real-world emotion detection task

🎨 Image Colorization with Pix2Pix

Transform black & white images to color automatically using conditional GANs!

📁 Location: 10-Pix2Pix/colorAI.ipynb
🎯 Technique: Conditional GAN, U-Net architecture
💾 Dataset: Paired grayscale/color images

🤖 RAG with Ollama

Build a document-aware chatbot that answers questions based on your PDFs!

📁 Location: 13-RAG/OllmaaOnColab.ipynb
🎯 Technique: Retrieval Augmented Generation, LLM
💾 Includes: Turkish document processing

😊 Facial Expression Recognition + Architecture Comparison

Detect emotions from facial images and compare 6 different architectures!

📁 Location: 5-ImageClass/CNN_FER.ipynb
📁 Advanced: 5-ImageClass/ArchitecturesAdv/architec.ipynb
🎯 Techniques: CNN, ResNet, DenseNet, EfficientNet, SE Attention, ViT
💾 Dataset: FER2013 (48x48 grayscale emotion faces, 7 classes)
🔬 Comparison: Parameters, FLOPs, training time, validation accuracy

🎯 YOLOv11 Object Detection

State-of-the-art real-time object detection and localization!

📁 Location: 6-ObjectDetection/roboflowAndYolov11/
🎯 Technique: YOLO architecture, Transfer learning
💾 Custom dataset support via RoboFlow

🛠️ Datasets Included

🌸 Iris Dataset - Classic ML dataset for classification
🚗 Car/Plane Images - Binary classification challenge
🐱🐶 Cats & Dogs - Famous Kaggle competition
😊 FER2013 - Facial expressions (7 emotions)
🏪 Northwind Database - Business data analysis
📈 Flight Data - Time series prediction
👗 Fashion MNIST - Clothing image generation
🌊 CalCOFI - Ocean temperature & marine data
💰 Tips Dataset - Restaurant tipping analysis
📸 Custom ImageNet subsets - Transfer learning

🎯 Prerequisites

Knowledge Requirements

✅ Basic Python programming
✅ High school mathematics (algebra, basic calculus helpful)
✅ Curiosity and patience! 🚀

Software Requirements

🐍 Python 3.8 or higher
📓 Jupyter Notebook / JupyterLab
🌐 Web browser for Google Colab
💾 Git for cloning the repository

Optional but Recommended

🎮 GPU for faster training (or use Google Colab)
💻 8GB+ RAM for local execution
🔧 Basic terminal/command line knowledge

🎓 How to Use This Repository

For Students 👨‍🎓👩‍🎓

Follow the numbered folders sequentially
Read theory notebooks before diving into code
Experiment with hyperparameters
Try modifying datasets and see what happens
Complete each project before moving to the next

For Educators 👨‍🏫👩‍🏫

Use as course material or supplementary content
Notebooks are ready for classroom presentation
Mix and match topics based on curriculum needs
Turkish content available for Turkish-speaking students
All code is well-commented and educational

For Practitioners 👨‍💻👩‍💻

Quick reference for implementing specific techniques
Copy and adapt code for your projects
Use datasets as templates for your own data
Benchmark your models against provided examples

⚡ What's New & Hot

🆕 Complete Model Zoo: Compare 6 architectures side-by-side on FER2013!
🔮 Vision Transformers: Learn the latest in image classification
🎯 Attention Mechanisms: Squeeze-and-Excitation blocks explained
📊 Scientific Benchmarking: Not just accuracy - FLOPs, params, and time!

💡 Pro Tips

🚀 Performance Tips

Always use GPU when available (Colab: Runtime → Change runtime type → GPU)
Start with small epochs to test your pipeline, then increase
Save checkpoints regularly during long training sessions
Use data augmentation to improve model generalization

🐛 Debugging Tips

Print tensor shapes frequently during development
Visualize your data before training
Start with a small batch size to catch errors early
Check for NaN losses - indicates learning rate or data issues

📊 Best Practices

Always split your data: Train/Validation/Test
Monitor both training and validation metrics
Use callbacks: EarlyStopping, ModelCheckpoint, ReduceLROnPlateau
Document your experiments and results

🌟 Contributing

Contributions are welcome! Whether it's:

🐛 Bug fixes
📝 Documentation improvements
✨ New tutorial additions
🌍 Translations
💡 Better explanations

How to Contribute:

Fork the repository
Create your feature branch (git checkout -b feature/AmazingTutorial)
Commit your changes (git commit -m 'Add amazing tutorial')
Push to the branch (git push origin feature/AmazingTutorial)
Open a Pull Request

📞 Contact & Support

📧 Issues: Open an issue on GitHub for bugs or questions
🌟 Stars: If you find this helpful, give it a star!
🔄 Share: Help others discover this resource

📄 License

This repository is created for educational purposes. Feel free to use the code for learning and teaching.

🙏 Acknowledgments

🎓 Built with passion for AI education
📚 Inspired by leading AI/ML courses and research
🌍 Community-driven and continuously improving
💻 Powered by open-source frameworks and tools

📈 Repository Stats

📂 14+ Major Topics
📓 50+ Jupyter Notebooks
🎯 10+ Different Datasets
🏗️ 6 Architecture Comparison (CNN, ResNet, DenseNet, EfficientNet, SE, ViT)
💻 Multiple ML/DL Frameworks (TensorFlow, PyTorch, scikit-learn)
🌐 Colab Ready Projects
🇹🇷 Bilingual Content (Turkish & English)
🔬 Scientific Methodology (FLOPs, params, benchmarking)

🎯 Ready to Start Your AI Journey?

Pick a topic, open a notebook, and start learning! 🚀

⭐ Don't forget to star this repo if you find it helpful! ⭐

Made with ❤️ for AI learners everywhere

🗺️ Quick Navigation

Beginner	Intermediate	Advanced
Prerequisites	Image Classification	🏆 Architecture Zoo
ML Basics	Object Detection	GANs
Theory	LSTM	Pix2Pix
Classification	Text Classification	RAG

Happy Learning! 🎓 Let's build amazing AI together! 🤖