Machine Learning on a Cancer Dataset - Part 22
This is the 4th tutorial on neural networks with scikit-learn.
In this video tutorial we take a look at the parameters that can be adjusted for multi-layer perceptrons in order to improve their performance. And they are many...
In the last video, we trained the neural network on the cancer dataset and got an improved performance as we've scaled the data and modified the number of iterations from 200 to 1000. To further improve the performance, here we're adjusting the regularization of the algorithm. Specifically, we're strengthening the regularization by increasing the alpha from 0.0001 to 1, which is pretty 'drastic'.
Additionally, we look at a nice and convenient way to access documentation for different libraries in the notebook itself. See the video for the full scoop.
As a reminder:
In this series I'm going to explore the cancer dataset that comes pre-loaded with scikit-learn. The purpose is to train the classifiers on this dataset, which consists of labeled data: ~569 tumor samples, each labeled malignant or benign, and then use them on new, unlabeled data.
Previous videos in this series: