Machine Learning with Imbalanced Data
5.0 /5- Offered byUDEMY
Machine Learning with Imbalanced Data at UDEMY Overview
Duration | 11 hours |
Total fee | ₹3,099 |
Mode of learning | Online |
Official Website | Go to Website  |
Credential | Certificate |
Machine Learning with Imbalanced Data at UDEMY Highlights
- Certificate of completion
- 11.5 hours on-demand video
- 20 articles
- 2 downloadable resources
- Access on mobile and TV
- 30-Day Money-Back Guarantee
- Full Lifetime Access
Machine Learning with Imbalanced Data at UDEMY Course details
- For Data scientists and machine learning engineers working with imbalanced datasets
- For Data scientists who want to improve the performance of models trained on imbalanced datasets
- For students who want to learn intermediate content on machine learning
- For students working with imbalanced multi-class targets
- Apply random under-sampling to remove observations from majority classes
- Perform under-sampling by removing observations that are hard to classify
- Carry out under-sampling by retaining observations at the boundary of class separation
- Apply random over-sampling to augment the minority class
- Create syntethic data to increase the examples of the minority class
- Implement SMOTE and its variants to synthetically generate data
- In this course, learner will learn multiple techniques which can use with imbalanced datasets to improve the performance of your machine learning models
- By the end of the course, learner will be able to decide which technique is suitable for dataset, and / or apply and compare the improvement in performance returned by the different methods on multiple datasets
Machine Learning with Imbalanced Data at UDEMY Curriculum
Introduction
Course Curriculum Overview
Course Material
Code
Jupyter notebooks
Presentations covered in the course
Python package Imbalanced-learn
Download Datasets
Additional resources for Machine Learning and Python programming
Machine Learning with Imbalanced Data: Overview
Imbalanced classes - Introduction
Nature of the imbalanced class
Approaches to work with imbalanced datasets - Overview
Additional Reading Resources (Optional)
Evaluation Metrics
Introduction to Performance Metrics
Accuracy
Accuracy - Demo
Precision, Recall and F-measure
Install Yellowbrick
Precision, Recall and F-measure - Demo
Confusion tables, FPR and FNR
Confusion tables, FPR and FNR - Demo
Balanced Accuracy
Balanced accuracy - Demo
Geometric Mean, Dominance, Index of Imbalanced Accuracy
Geometric Mean, Dominance, Index of Imbalanced Accuracy - Demo
ROC-AUC
ROC-AUC - Demo
Precision-Recall Curve
Precision-Recall Curve - Demo
Comparison of ROC and PR curves - Optional
Additional reading resources (Optional)
Probability
Metrics for Mutliclass
Metrics for Multiclass - Demo
PR and ROC Curves for Multiclass
PR Curves in Multiclass - Demo
ROC Curve in Multiclass - Demo
Udersampling
Under-Sampling Methods - Introduction
Random Under-Sampling - Intro
Random Under-Sampling - Demo
Condensed Nearest Neighbours - Intro
Condensed Nearest Neighbours - Demo
Tomek Links - Intro
Tomek Links - Demo
One Sided Selection - Intro
One Sided Selection - Demo
Edited Nearest Neighbours - Intro
Edited Nearest Neighbours - Demo
Repeated Edited Nearest Neighbours - Intro
Repeated Edited Nearest Neighbours - Demo
All KNN - Intro
All KNN - Demo
Neighbourhood Cleaning Rule - Intro
Neighbourhood Cleaning Rule - Demo
NearMiss - Intro
NearMiss - Demo
Instance Hardness - Intro
Instance Hardness Threshold - Demo
Instance Hardness Threshold Multiclass Demo
Undersampling Method Comparison
Wrapping up the section
Setting up a classifier with under-sampling and cross-validation
Summary Table
Oversampling
Over-Sampling Methods - Introduction
Random Over-Sampling
Random Over-Sampling - Demo
ROS with smoothing - Intro
ROS with smoothing - Demo
SMOTE
SMOTE - Demo
SMOTE-NC
SMOTE-NC - Demo
SMOTE-N
SMOTE-N Demo
ADASYN
ADASYN - Demo
Borderline SMOTE
Borderline SMOTE - Demo
SVM SMOTE
Resources on SVMs
SVM SMOTE - Demo
K-Means SMOTE
K-Means SMOTE - Demo
Over-Sampling Method Comparison
Wrapping up the section
How to Correctly Set Up a Classifier with Over-sampling
Setting Up a Classifier - Demo
Summary Table
Over and Undersampling
Combining Over and Under-sampling - Intro
Combining Over and Under-sampling - Demo
Comparison of Over and Under-sampling Methods
Combine over and under-sampling manually
Wrapping up
Ensemble Methods
Ensemble methods with Imbalanced Data
Foundations of Ensemble Learning
Bagging
Bagging plus Over- or Under-Sampling
Boosting
Boosting plus Re-Sampling
Hybdrid Methods
Ensemble Methods - Demo
Wrapping up
Additional Reading Resources
Cost Sensitive Learning
Cost-sensitive Learning - Intro
Types of Cost
Obtaining the Cost
Cost Sensitive Approaches
Misclassification Cost in Logistic Regression
Misclassification Cost in Decision Trees
Cost Sensitive Learning with Scikit-learn
Find Optimal Cost with hyperparameter tuning
Bayes Conditional Risk
MetaCost
MetaCost - Demo
Optional: MetaCost Base Code
Additional Reading Resources
Probability Calibration
Probability Calibration
Probability Calibration Curves
Probability Calibration Curves - Demo
Brier Score
Brier Score - Demo
Under- and Over-sampling and Cost-sensitive learning on Probability Calibration
Calibrating a Classifier
Calibrating a Classifier - Demo
Calibrating a Classfiier after SMOTE or Under-sampling
Calibrating a Classifier with Cost-sensitive Learning
Probability: Additional reading resources
Putting it all together
Examples
Next steps
Vote for the next course!
Congratulations
Bonus Lecture
Other courses offered by UDEMY
Machine Learning with Imbalanced Data at UDEMY Students Ratings & Reviews
1 Rating- 4-51
Student Forum
Anything you would want to ask experts?
Data ScienceMachine LearningMachine Learning and Pattern RecognitionMachine Learning with Imbalanced Data