Machine Learning Study Timetable (6 Months)
This timetable is structured to help you complete a comprehensive machine learning
curriculum in 6 months. It includes dedicated Sundays for revision to reinforce your
learning.
Month 1: Foundations of Machine Learning (4 weeks)
Week 1: The Machine Learning Landscape
Topics (Monday–Saturday):
- What is Machine Learning?
- Types of Learning: Supervised/Unsupervised Learning, Batch/Online Learning.
- Challenges in ML: Overfitting, underfitting, data quality issues.
Practical: Implement a basic supervised learning model (e.g., Linear Regression in Scikit-
learn).
Sunday Revision:
- Review types of learning, ML challenges, and Scikit-learn basics.
- Practice quizzes or small exercises.
Week 2-3: End-to-End Machine Learning Project
Topics (Monday–Saturday):
- Data Preprocessing: Cleaning, scaling, pipelines.
- Exploratory Data Analysis (EDA): Visualizations, correlation analysis.
- Train-Test Split and Cross-Validation.
Practical: Perform EDA and preprocessing on a dataset. Build your first pipeline.
Sunday Revision:
- Review key steps in EDA, preprocessing techniques, and cross-validation.
- Revise code written during the week.
Week 4: Classification
Topics (Monday–Saturday):
- Binary Classification (Logistic Regression), performance metrics (Precision, Recall, ROC).
- Multiclass, multilabel classification.
Practical: Train classifiers and evaluate models using metrics.
Sunday Revision:
- Review confusion matrix, precision-recall tradeoff.
- Practice using Scikit-learn's classification metrics.
Month 2: Core Algorithms and Models (4 weeks)
Week 1-2: Training Models
Topics (Monday–Saturday):
- Linear Regression, gradient descent, regularization techniques (Ridge, Lasso).
- Logistic Regression for classification.
Practical: Implement gradient descent and train linear models.
Sunday Revision:
- Review gradient descent variations and regularization.
- Solve math problems related to cost functions and optimization.
Week 3: Support Vector Machines (SVMs)
Topics (Monday–Saturday):
- Linear and kernel SVMs, soft margin classification.
Practical: Train and visualize SVMs on toy datasets using Scikit-learn.
Sunday Revision:
- Review SVM concepts and kernel tricks.
- Revise practical code for SVMs.
Week 4: Decision Trees and Random Forests
Topics (Monday–Saturday):
- Decision Trees (Gini impurity, entropy), ensemble methods (Bagging, Boosting, Random
Forests).
Practical: Train and analyze Decision Trees and Random Forests.
Sunday Revision:
- Review Decision Trees' strengths and weaknesses, feature importance in Random Forests.
Month 3: Advanced ML Techniques (4 weeks)
Week 1-2: Dimensionality Reduction
Topics (Monday–Saturday):
- PCA, kernel PCA, and applications in data compression and visualization.
Practical: Apply PCA on datasets and interpret variance explained.
Sunday Revision:
- Review curse of dimensionality and PCA workflows.
Week 3-4: Unsupervised Learning
Topics (Monday–Saturday):
- Clustering (K-Means, DBSCAN), Gaussian Mixture Models, anomaly detection.
Practical: Implement clustering techniques for segmentation tasks.
Sunday Revision:
- Review strengths and weaknesses of clustering techniques.
- Practice clustering on unseen datasets.
Month 4: Neural Networks Basics (4 weeks)
Week 1-2: Introduction to Neural Networks with Keras
Topics (Monday–Saturday):
- Artificial neurons, MLPs, backpropagation.
- Training and evaluating neural networks for classification and regression.
Practical: Build MLPs using Keras.
Sunday Revision:
- Review neural network architecture and hyperparameter tuning basics.
Week 3-4: Training Deep Neural Networks
Topics (Monday–Saturday):
- Vanishing gradients, optimizers (Adam, RMSProp).
- Regularization techniques (dropout, early stopping).
Practical: Experiment with optimizers and regularization in Keras.
Sunday Revision:
- Review regularization and optimizer choices.
- Revise model training workflows.
Month 5: Deep Learning Applications (4 weeks)
Week 1-2: Convolutional Neural Networks (CNNs)
Topics (Monday–Saturday):
- CNN layers (convolutional, pooling), architectures (ResNet, VGG).
- Transfer learning with pretrained models.
Practical: Train a CNN for image classification using Keras.
Sunday Revision:
- Review CNN concepts and architecture comparisons.
Week 3-4: Data Handling with TensorFlow
Topics (Monday–Saturday):
- Data pipelines, TFRecords, feature columns.
Practical: Build efficient data workflows using TensorFlow.
Sunday Revision:
- Review TensorFlow Data API and preprocessing techniques.
Month 6: Projects and Special Topics (4 weeks)
Week 1-2: Projects
Topics (Monday–Saturday):
- Apply all concepts to build a comprehensive project (classification, regression, or
clustering).
Sunday Revision:
- Review end-to-end ML project workflows, debugging, and optimization.
Week 3: General Revision
Revise foundational ML topics (gradient descent, regularization, classification metrics).
Week 4: Advanced Exploration
Explore specific topics like GANs, RNNs, or reinforcement learning.
