Machine Learning with Python
Learn Machine Learning with Python, a comprehensive training manual that teaches the fundamentals of coding with Python. Whether you want to improve your coding skills or you want an upgrade at your workplace, this is the ideal start. In this course, you’ll master the processes, patterns, and strategies of this user-friendly programming language. This Python ML course covers supervised learning paradigms, like classical algorithms, and regression techniques to evaluate performance metrics. Besides this, you’ll also learn feature engineering for converting raw data into meaningful features. Furthermore, you’ll also leverage the Python scikit-learn library along with other powerful tools.
Practice on our Labs to solidify your understanding as you explore object-oriented programming, modules, error handling, and even file operations. By the end of this course, you’ll be confidently writing Python ML scripts and resolving coding issues.
What You`ll Learn on This Course
-
- Understand the fundamentals of supervised machine learning algorithms and their classification
- Evaluating performance metrics for assessing the efficacy of your models
- Using engineer features to convert raw data into meaningful ML algorithms
- Managing system performance by creating robust pipelines
- Apply ML to various data types
- Leverage Python scikit-learn library and other tools
- Use of advanced techniques like neural networks and graphical models
This course comes with:
Curriculum
- 2 Sections
- 0 Lessons
- 26 Weeks
- Lesson Plan
Introduction to Machine Learning
- Welcome, Scope, Terminology, Prediction, and Data
- Examples of Learning Systems and Evaluation
- Building Learning Systems: Process, Assumptions, and Reality
Technical Foundations
- Mathematical Background: Probability, Linear Algebra, Dot Products
- Geometric View of Data, Notation, Nonlinearity, Floating-Point Issues
- Software and Setup for Machine Learning
Classification
- Predicting Categories: Simple Datasets and Classifiers
- Evaluation Metrics, ROC, Precision-Recall, Lift Curves
- Advanced Classification Methods: Decision Trees, SVM, Logistic Regression, Discriminant Analysis
Regression
- Predicting Numerical Values: Nearest Neighbors, Linear Regression
- Evaluating Regressors: Residuals, Standardization, Advanced Metrics
- Advanced Regression Methods: Regularization, Support Vector Regression, Regression Trees
Model Evaluation and Comparison
- Bias, Variance, Overfitting/Underfitting, Cross-Validation
- Comparing Learners and Sophisticated Evaluation Techniques
Feature Engineering
- Manual Feature Engineering: Selection, Scaling, Discretization, Categorical Coding, Interactions
- Domain-Specific Features: Text, Images, Clustering
- Feature Selection and Construction: PCA, Kernels
Model Tuning and Pipelines
- Hyperparameter Tuning and Nested Cross-Validation
- Pipelines and Combining Pipelines with Tuning
Ensemble Methods
- Voting, Bagging, Random Forests, Boosting
- Comparing Tree-Ensemble Methods
Advanced Modeling Concepts
- Optimization Techniques
- Building Models from Raw Materials: Linear Regression, Logistic Regression, SVM, Neural Networks
- Probabilistic Graphical Models
Appendices
- Appendix A: mlwpy.py Listing
0 - Hands-on LAB Activities
Technical Foundations
- Plotting Graphs: Probability Distribution, Line, 3D, Polynomial
- Using Python Functions: zip(), numpy.dot(), sum of squares
Classification
- Displaying Histograms
- Confusion Matrix, ROC Curves, Logistic Model Evaluation
- Working with Datasets: load_digits()
Regression
- Handling Outliers and Median Calculation
- Estimating Regression Equations
- Scatterplots, Trendlines, Piecewise Constant Regression
Model Evaluation and Comparison
- Variance, Standard Deviation, Prediction Error Rates
- Swarm Plots and Summary Statistics
Feature Engineering
- Manual Feature Engineering: Manipulating Inputs and Targets
- Automatic Feature Engineering: Correlation Matrix, Nonlinear Models, PCA, Manifold Methods
- Domain-Specific Feature Engineering: Text Encoding
Combining Learners
- Calculating Means and Creating Ensemble Models
Extensions and Further Directions
- Building an Estimated Simple Linear Regression Equation
0
