Dimensionality Reduction & Feature Extraction Techniques in Python: Part two

Part two of the Dimensionality Reduction & Feature Extraction goes over the real-world application of using these techniques. Download the file by clicking here


  • Anaconda Navigator / Jupyter notebook
  • Keras, ipywidgets, seaborn

How to use:

  • Compile the first 6 Cells
    • Stop on the cell starting with the code: display(test_size)
  • The last cell should generate a bunch of User Interfaces that can allow you to set the parameters you desire
  • Click run me to run the Deep Neural Network


This Jupyter notebook will give you the chance to run the same Sequential Neural Network on the iris dataset with different reduction techniques, epochs, validation/verification split, and batch sizes. These parameters can affect the overall training time and accuracy of a neural network and I encourage you to find which set of parameters take the longest, and shortest training time!

Share this Post

Leave a Comment