Grid Search for Mini Batch Gradient Descent
This report contains the run for the grid search of different mini-batch sizes and learning rates. We have the following list for the two hyperparameters
lr_list=[0.0001, 0.001, 0.01, 0.1]
batch_size_list = [1,8,16,32,64]
Created on February 14|Last edited on February 14
Comment
Section 1
This set of panels contains runs from a private project, which cannot be shown in this report
The learning curve for different configurations. You can zoom in on individual plots to see how they are performing
This set of panels contains runs from a private project, which cannot be shown in this report
This is the grid search run for mini-batch gradient descent. We can see that for a higher batch size, we have less total time taken, and for the lower batch size we have higher time.
This set of panels contains runs from a private project, which cannot be shown in this report
Add a comment