Uploaded August 2018 | Updated September 2026, 2 weeks ago
Watch the full series on deep neural networks (to date): youtube.com/watch?v=9-QYsN_knG4&list=PLsu0TcgLDUiIKPMXu1k_rItoTV8xPe1cj
In the preceding video ( youtube.com/watch?v=CHJzoFArI8c&list=PLsu0TcgLDUiIKPMXu1k_rItoTV8xPe1cj&index=10 ) we created a deep neural network using Keras in R. The model provided error values for the training and the validation set.
In this video I discuss the use of these errors to diagnose bias and variance in the model. These errors require correction.
I also discuss issues around the training and test sets as well as the concept of the ground truth.
The RPubs document is available at rpubs.com/juanhklopper/poor_performance_of_a_neural_network_model
The RStudio file is available on Github at github.com/juanklopper/Deep-learning-using-R
Watch the full series on deep neural networks (to date): youtube.com/watch?v=9-QYsN_knG4&list=PLsu0TcgLDUiIKPMXu1k_rItoTV8xPe1cj
In the preceding video ( youtube.com/watch?v=CHJzoFArI8c&list=PLsu0TcgLDUiIKPMXu1k_rItoTV8xPe1cj&index=10 ) we created a deep neural network using Keras in R. The model provided error values for the training and the validation set.
In this video I discuss the use of these errors to diagnose bias and variance in the model. These errors require correction.
I also discuss issues around the training and test sets as well as the concept of the ground truth.
The RPubs document is available at rpubs.com/juanhklopper/poor_performance_of_a_neural_network_model
The RStudio file is available on Github at github.com/juanklopper/Deep-learning-using-R
