Supervised Machine Learning Using Python

• Subsection 1.1.1: What is Machine Learning?
  • Definition and categories (Supervised, Unsupervised, Reinforcement Learning)
• Subsection 1.1.2: Key Concepts in Supervised Learning
  • Training and testing
  • Features and labels

• Subsection 1.2.1: Common Use Cases
  • Email spam detection
  • Predicting house prices
  • Customer churn analysis
• Subsection 1.2.2: Benefits and Limitations

• Subsection 1.3.1: Installing Python and Jupyter Notebook
• Subsection 1.3.2: Key Libraries for Machine Learning
  • NumPy, Pandas, Matplotlib, Scikit-learn

• Subsection 2.1.1: Loading Data with Pandas
• Subsection 2.1.2: Handling Missing Values
  • Imputation techniques (mean, median, mode)

• Subsection 2.2.1: Descriptive Statistics
  • Mean, median, standard deviation
• Subsection 2.2.2: Data Visualization
  • Histograms, scatter plots, correlation heatmaps

• Subsection 2.3.1: Encoding Categorical Data
  • One-hot encoding, label encoding
• Subsection 2.3.2: Feature Scaling
  • Standardization and normalization

• Subsection 3.1.1: Linear Regression
  • Concept and mathematical representation
  • Training and evaluating a linear regression model using Scikit-learn
• Subsection 3.1.2: Polynomial Regression
  • When and why to use polynomial regression
  • Implementing polynomial regression in Python

• Subsection 3.2.1: Logistic Regression
  • Concept and sigmoid function
  • Building and evaluating a logistic regression model
• Subsection 3.2.2: k-Nearest Neighbors (k-NN)
  • How k-NN works (distance metrics, k-value selection)
  • Implementing k-NN for classification tasks

• Subsection 4.1.1: Decision Trees
  • How decision trees work (splitting criteria, entropy, and Gini index)
  • Building and visualizing decision trees in Python
• Subsection 4.1.2: Random Forests
  • Ensemble learning and bagging
  • Training a random forest model using Scikit-learn

• Subsection 4.2.1: Evaluation Metrics
  • Regression: Mean Squared Error (MSE), R-squared
  • Classification: Accuracy, Precision, Recall, F1-Score, ROC-AUC
• Subsection 4.2.2: Cross-Validation
  • Concept of k-Fold Cross-Validation
  • Implementing cross-validation for robust evaluation

• Subsection 4.3.1: Grid Search
• Subsection 4.3.2: Random Search

• Subsection 5.1.1: Saving and Loading Models
  • Using joblib and pickle
• Subsection 5.1.2: Integrating Models into Applications
  • Simple API deployment using Flask

• Subsection 5.2.1: Problem Definition
  • Example: Predicting customer churn for a telecom company
• Subsection 5.2.2: Model Development
  • Data preprocessing, training, evaluation
• Subsection 5.2.3: Deployment and Presentation
  • Building a basic user interface to showcase predictions

• • Goal: Apply everything learned in the course to solve a real-world supervised learning problem.
  • Steps:
    • Data preprocessing
    • Model selection and evaluation
    • Hyperparameter tuning
    • Deployment of the final model
  • Outcome: A portfolio-worthy project demonstrating your expertise in supervised machine learning.

We accept 20+ global currencies. View list →

Take your research to the next level with NanoSchool.