Practical, project-driven curriculum with weekly assignments and mini-projects to build production-ready ML competencies.
WEEK 1 — Introduction to AI & ML
- Concepts: AI, ML, DL, Data Science overview
- Types of ML: Supervised, Unsupervised, Reinforcement
- Real-world applications (healthcare, finance, retail)
- Setting up environment: Anaconda / Jupyter / Google Colab
- Outcome: Understand AI ecosystem and setup environment
- Tools: Jupyter Notebook, Google Colab
WEEK 2 — Python Programming for ML
- Python syntax, data types, loops, functions
- Lists, tuples, sets, dictionaries
- File handling, exception handling
- Assignment: Write a Python script to analyze student grades
WEEK 3 — Data Handling with NumPy & Pandas
- NumPy arrays, broadcasting, slicing
- Pandas: reading CSV/Excel, merging, grouping, cleaning
- Handling missing and duplicate data
- Mini-project: Analyze a dataset (sales or COVID data)
WEEK 4 — Data Visualization
- Matplotlib and Seaborn for charts
- Line, bar, pie, histogram, scatter, heatmaps
- Plot styling and dashboards
- Assignment: Visualize trends using any public dataset
WEEK 5 — Mathematics & Statistics for ML
- Descriptive statistics (mean, variance, std. deviation)
- Probability distributions
- Correlation & covariance
- Hypothesis testing and p-values
WEEK 6 — ML Fundamentals
- ML process: data → model → evaluation
- Train/Test split, random state
- Linear regression from scratch
- Assignment: Predict house prices using Linear Regression
WEEK 7 — Regression Models
- Multiple Linear, Polynomial, Ridge & Lasso regression
- Model performance metrics (R², RMSE, MAE)
WEEK 8 — Classification Models
- Logistic Regression, KNN
- Decision Tree, Random Forest
- Confusion Matrix, ROC Curve, Precision-Recall
- Mini-project: Predict if a customer will buy a product
WEEK 9 — Feature Engineering
- Data encoding (Label, One-hot)
- Outlier detection and removal
- Feature scaling: StandardScaler, MinMaxScaler
WEEK 10 — Model Optimization
- Overfitting vs Underfitting
- Cross-validation (K-Fold)
- GridSearchCV, RandomizedSearchCV
- Assignment: Tune hyperparameters for a Random Forest model
WEEK 11 — Unsupervised Learning
- K-Means Clustering
- Hierarchical clustering
- PCA for dimensionality reduction
WEEK 12 — Ensemble Techniques
- Bagging & Boosting: Random Forest, XGBoost, AdaBoost
- Mini-project: Customer segmentation using clustering
WEEK 13 — Neural Network Basics
- Neuron, weights, bias, activation functions
- Forward propagation, loss, gradient descent
WEEK 14 — Building ANN Models
- TensorFlow & Keras basics
- Creating ANN for binary/multi-class classification
- Dropout and regularization
- Assignment: Build an ANN for bank loan approval
WEEK 15 — Convolutional Neural Networks (CNNs)
- Convolution, pooling, flattening layers
- CNN architecture for image classification
WEEK 16 — Transfer Learning
- Pretrained models: VGG16, ResNet, MobileNet
- Mini-project: Image classification using CNN
WEEK 17 — NLP Basics
- Text preprocessing (tokenization, stemming, lemmatization)
- Stopwords, Bag of Words, TF-IDF
- NLTK, SpaCy libraries
WEEK 18 — Sentiment Analysis
- Text classification with Naive Bayes & Logistic Regression
- Word2Vec, GloVe embeddings
- Mini-project: Sentiment analysis on tweets
WEEK 19 — Advanced NLP
- RNN, LSTM, GRU concepts
- Transformers: BERT, GPT overview
WEEK 20 — Reinforcement Learning (RL)
- RL concepts: Agent, Environment, Reward
- Q-learning, Deep Q-Network
- Assignment: Simulate a simple RL agent
WEEK 21 — Model Deployment
- Save models (Pickle/Joblib)
- Flask / FastAPI model APIs
- Streamlit dashboard for model visualization
WEEK 22 — MLOps Introduction
- Version control (Git, GitHub)
- Model tracking (MLflow)
- Intro to Docker for containerization
WEEKS 23–24 — Capstone Project & Presentation
- Choose one project:
- 1. Customer churn prediction
- 2. Image classification web app
- 3. Sales forecasting dashboard
- 4. Sentiment analysis using LSTM
- Deliverables: Source code (GitHub), Streamlit/Flask app demo, Project presentation