AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |
Back to Blog
![]() ![]() Although this is often used to demonstrate classical ML algorithms (e.g., boosting, decision trees), neural networks can also be applied to this style of dataset. Starting Simple - The Iris Dataset (Tabular)Ī very common dataset to test algorithms with is the Iris Dataset, a simple 4-attribute classification dataset. Most models from the Model Zoo have the added benefit of a pre-trained option, which tends to improve generality and reduce the training required on your end to achieve good performance.Īll datasets/models referenced in this tutorial can be found in the asset pack. Use a well-known pre-defined architecture from the Model Zoo. Design your own architecture, specifying a custom layer setup WekaDeeplearning4j allows you to do this in one of two ways, both of which will be explained in this section: ![]() ![]() This section walks through a common deep learning task - training a Neural Network. ![]()
0 Comments
Read More
Leave a Reply. |