Learn to create powerful ensemble models using random forests, deep neural networks, and other techniques in TensorFlow and Keras. Explore the implementation of ensembles using Keras and scikit-learn libraries in this 24-minute video tutorial. Discover how to evaluate model performance using log loss, assess feature importance, and apply perturbation ranking. Gain insights into classification evaluation, error analysis, and the application of these techniques to biological and chemical response data. Follow along with hands-on neural network training examples and explore advanced concepts like perturbation rank columns and Samba. This video is part of a comprehensive deep learning course offered at Washington University in St. Louis, providing valuable knowledge for both students and self-learners interested in advanced machine learning techniques.
Overview
Syllabus
Introduction
Log Loss
Log Loss Output
Feature Importance
Perturbation Ranking
Perturbation Explained
Classification
Evaluation
Error vs Importance
Biological Response Data
Chemical Response Data
Neural Network Training
Perturbation Rank
Columns
Samba
Taught by
Jeff Heaton
Related Courses
-
Introduction to Neural Networks with TensorFlow
-
Keras with TensorFlow Course - Python Deep Learning and Neural Networks for Beginners Tutorial
-
Journey into Machine Learning with Sklearn and Tensorflow
-
Deep Learning with Keras and Practical Applications
-
Advanced Generative Adversarial Networks (GANs)
-
Applications of Deep Neural Networks for TensorFlow and Keras
