🤖 Introduction to Artificial Intelligence

This repository contains the material for the Introduction to Artificial Intelligence course aimed at undergraduate students. The course introduces the theoretical and practical foundations of Machine Learning and Artificial Intelligence through guided tutorials, model-focused exercises, and real-world applications.

---

🧭 Course Structure

The course is organized into thematic units, each including:

• 📘 Tutorials: End-to-end machine learning pipelines for realistic problems.
• 🔬 Practicals: Focused exercises for deepening understanding of specific models.
• 🌐 Web Applications: Deployment of trained models using Flask.

---

📚 Units

✅ Unit 1: Fundamentals, Linear Models, and Dimensionality Reduction

• Topics: risk minimization, generalization, underfitting, overfitting
• Models: linear regression, logistic regression, Principal Component Analysis (PCA)
• Extras: model deployment via Flask

Notebooks:

• simple_pipeline.ipynb: Our first pipeline to understand some key concepts: generalization, overfitting, underfitting, error minimization, etc.
• guia_1.ipynb: Introductory pipeline with housing prices dataset
• guia_1_extended.ipynb: Extended version with additional features and improved model
• guia_2.ipynb: Classification pipeline using astronomical data
• practico_1.ipynb: Regression analysis on global temperature data
• practico_2.ipynb: Logistic regression for university admission prediction
• practico_3.ipynb: PCA on handwritten digits
• app.py: Web application with a deployed regression model

Datasets:

Notebook	Dataset
guia_1.ipynb, extended	Used Houses RM Chile (May 2020) Chilean cities
guia_2.ipynb	Stellar Classification SDSS17
practico_1.ipynb	Global Temperature Records
practico_2.ipynb	University Admission Chile
practico_3.ipynb	Digit Recognizer (MNIST)

---

🧠 Unit 2: Clustering and Non-Parametric Classification Models

• Unsupervised Learning: K-Means, Agglomerative Clustering, DBSCAN, Gaussian Mixtures
• Supervised Learning: Naive Bayes, K-Nearest Neighbors (KNN), Decision Trees, Random Forest
• Bayesian Reasoning: Illustrations of the sequential nature of Bayes' theorem

Notebooks:

• guia_1.ipynb: Clustering with four algorithms on the penguin dataset
• bayes_1.ipynb, bayes_2.ipynb: Bayes theorem applied to toy examples
• practico_1.ipynb: Naive Bayes for tweet sentiment classification
• practico_2.ipynb: KNN classifier to predict Alzheimer's status
• practico_3.ipynb: KNN, decision tree and random forest music genre classifier
• practico_4.ipynb: KNN, decision tree and random forest music genre classifier

Datasets:

Notebook	Dataset
guia_1.ipynb	Clustering Penguins
practico_1.ipynb	Sentimental Analysis for Tweets
practico_2.ipynb	Alzheimer Features
practico_3.ipynb	Music features
practico_4.ipynb	Spam Text Message Classification

---

🧠 Unit 3: Deep Learning

• Multilayer Perceptron
• Transfer learning and Fine tuning

Notebooks:

• practico_1.ipynb: Use a model, apply transfer learning and finetuning.
• practico_2.ipynb: Basic concepts of dense neural networks.

Datasets:

Notebook	Dataset
practico_1.ipynb	Cats and Dogs
practico_2.ipynb	Simulated data
practico_3_CNN.ipynb	CIFAR-10 (imported directly from the notebook)

---

⚙️ Requirements

Install Python 3 and the following libraries:

pip install numpy pandas scikit-learn matplotlib seaborn flask joblib nltk

Install torch following the instructions of Get Started

pip install install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

You may also want:

• Jupyter Notebook or JupyterLab
• VSCode
• venv or conda for environment management

---

🚀 How to Use This Repository

1. Clone the repository:

git clone https://github.com/GabrielCabas/ML_Course.git
cd ML_Course

2. Open the notebooks:

jupyter notebook

3. (Optional) Run web applications like app.py:

python unidad_1/app.py

---

🛠 Author and Contributions

This course was created by Gabriel Cabas for undergraduate students beginning their journey in Artificial Intelligence.
Contributions, feedback, and pull requests are welcome!

---

📜 License

This repository is released under the MIT License. You are free to use, modify, and share the content with appropriate attribution.