
Artificial Intelligence & Machine Learning (21 days Program Outline)
Week 1: Foundations of AI & Machine Learning (Days 1–7)
Day 1 – Introduction
- What is AI vs ML vs Deep Learning - Types of ML: Supervised, Unsupervised, Reinforcement Learning - AI applications - Tools: Python, Jupyter, scikit-learn, Pandas, Numpy
Day 2 – Python for ML
- Data types, loops, functions - Numpy for numerical data - Pandas for data handling - Matplotlib & Seaborn for visualization
Day 3 – Math for ML
- Linear algebra basics - Probability & stats (mean, variance, std dev, distributions) - Gradient descent intro
Day 4 – Data Preprocessing
- Handling missing data, outliers - Label encoding, one-hot encoding - Feature scaling (Normalization, Standardization) - Train/test split
Day 5 – Supervised Learning Overview
- Regression vs Classification - Model training/testing - Evaluation metrics (accuracy, precision, recall, F1)
Day 6 – Linear Regression
- Simple/multiple linear regression - Loss function, gradient descent - Hands-on: Predict house prices
Day 7 – Logistic Regression
- Binary classification - Sigmoid function - Hands-on: Classify diabetes dataset
Week 2: Core ML Algorithms (Days 8–14)
Day 8 – Decision Trees & Random Forests
- Tree building, overfitting - Ensemble methods - Hands-on: Titanic survival prediction
Day 9 – Support Vector Machines (SVM)
- Linear & kernel SVM - Hyperparameters - Hands-on: Classify digits
Day 10 – K-Nearest Neighbors (KNN)
- Lazy learning - Choosing K - Hands-on: Iris classification
Day 11 – Naive Bayes
- Bayes theorem - Types: Gaussian, Multinomial - Hands-on: Spam detection
Day 12 – Unsupervised Learning (Clustering)
- K-Means clustering - Elbow method - Hands-on: Customer segmentation
Day 13 – Dimensionality Reduction
- PCA (Principal Component Analysis) - t-SNE (for visualization) - Hands-on: Visualize MNIST data
Day 14 – Model Evaluation & Tuning
- Cross-validation - Hyperparameter tuning (GridSearch, RandomSearch) - Avoiding overfitting (regularization, dropout)
Week 3: Deep Learning & Projects (Days 15–21)
Day 15 – Intro to Deep Learning
- Neural networks - Perceptron, activation functions - Feedforward, backpropagation
Day 16 – TensorFlow/Keras Basics
- Building models in Keras - Compile, fit, evaluate - Hands-on: Binary classifier (Keras)
Day 17 – CNNs (Convolutional Neural Networks)
- Image data - Convolutions, pooling, filters - Hands-on: Classify MNIST/CIFAR10
Day 18 – RNNs (Recurrent Neural Networks)
- Sequence modeling - LSTM, GRU - Hands-on: Sentiment analysis
Day 19 – NLP with ML
- Text preprocessing - TF-IDF, Word2Vec - Hands-on: Text classification
Day 20 – Mini Project Day
Choose 1: - Stock price prediction (Regression) - Fake news detection (NLP) - Disease prediction (Classification) - Image classifier (CNN)
Day 21 – Final Wrap-up & Deployment
- Review concepts - Model deployment: Streamlit / Flask intro - Exporting ML models with Pickle - Next steps in AI (e.g., GPT, RL, MLOps)
🛠 Tools You’ll Use: – Python – Jupyter Notebook – scikit-learn – TensorFlow/Keras – Pandas, Numpy, Matplotlib – Google Colab (optional for GPU)
Schedule
