Machine Learning
Quality, not quantity
This 8-week program is designed to provide comprehensive theoretical and practical learning in Classical Machine Learning, with no prerequisites required. It covers end-to-end learning, focusing on the three major components: Supervised Learning, Unsupervised Learning, and Reinforcement Learning. Participants will delve into both regression and classification, starting with data extraction, moving through feature engineering and feature selection, and culminating in model training and evaluation. This structured approach ensures a thorough understanding of each step in the machine learning process.
Content
No Pre-requisites required
Mock Interview Included
Learn Classical ML in 8 weeks (2.5 Hours Each)
Week 1: Introduction to Machine Learning
Lecture:
Overview of Machine Learning
Types of Machine Learning: Supervised, Unsupervised, Reinforcement
Applications of Machine Learning in Engineering
Lab:
Setting up the environment (Python, Jupyter Notebooks, Libraries)
Introduction to basic Python for ML
Week 2: Data Preprocessing
Lecture:
Understanding Data: Types, Features, and Quality
Data Cleaning: Handling missing values, outliers
Data Transformation: Normalization, Standardization, Encoding categorical variables
Lab:
Practical exercises on data preprocessing using pandas and scikit-learn
Week 3: Supervised Learning - Regression
- Lecture:
Linear Regression: Concepts, Assumptions, and Implementation
Evaluation Metrics: MAE, MSE, RMSE, R²
Regularization techniques: Ridge, Lasso
- Lab:
Implementing Linear Regression models
Evaluating and tuning the models
Week 4: Supervised Learning - Classification
Lecture:
Evaluation Matrix
Confusion Matrix
F1 score, Precision, Recall
Qscore, ROC-AUC
Lab
Building and evaluating classification models
Week 5: Unsupervised Learning
Lecture:
Clustering: K-means, Hierarchical Clustering
Dimensionality Reduction: PCA, t-SNE
Lab:
Practical exercises on clustering and dimensionality reduction
Week 6: Model Evaluation and Selection
Lecture:
Cross-Validation techniques
Bias-Variance Tradeoff
Hyperparameter tuning: Grid Search, Random Search
Lab:
Applying cross-validation and hyperparameter tuning on models
Week 7: Advanced Topics
Lecture:
Ensemble Methods: Bagging, Boosting (Random Forest, Gradient Boosting)
Support Vector Machines (SVM)
Lab:
Implementing and evaluating ensemble methods and SVM
Week 8: Project and Review
Lecture:
Overview of a machine learning project workflow
Best practices in model deployment and maintenance
Lab:
Students work on a capstone project, applying all the concepts learned
Presentations and reviews
Additional Resources
Miscellaneous: Doubts session