Skip to content

kbmclaren/tensorFlow-CMSC478-ML

Repository files navigation

tensorFlow-CMSC478-ML

Description

The Tensorflow Assignment focuses on verifying the correct setup of Tensorflow2/Keras, and then using Tensorflow/Keras modules to build introductory neural networks

The Jupyter notebook in this repo is the notebook I submitted to be graded. This notebook was a guided application of ML techniques to build a One-Layer One-Neuron Regression Neural Network using Tensorflow/Keras, a Multi-Layer NN for Multi-Class Classification on MNIST Dataset, practice a new capability in Tensorflow2 callled "Eager Execution", create models via the Sequential API, apply these lessons to the MNIST Fashion database, create a regression model for California house prices, use callbacks, and practice visualization via Tensorboard.

The end results was that I was introduced to the many powerful and convenient features of Tensorflow2 and Keras, and practiced using these features by building four models of varying size and complexity.