Reference no: EM132600595 , Length: 7 pages
Predictive Modeling
Assignment 5 - Deep Learning
Deep Learning Assignment Overview
You'll learn to build a deep learning neural network, focusing on image recognition, sometimes called computer vision. This is one of the primary applications of deep learning neural networks. To do this you'll do four steps.
1. Read the background material on deep learning and on deep learning in Python.
2. Set up your system to be compatible with the required software.
3. Run through the example script with the example data.
4. Complete the "starter" scripts and write a paper summarizing your results.
Background on Deep Learning
Read the suggested sources in the classroom to gain familiarity with the subject matter.
Set up your system
You will be using Python 3 and several libraries (notably TensorFlow and Keras) to do this - review the course documents for instructions on downloading Python and setting up your environment:
Once your system is set up, download the example code and data files, which use the MNIST digits data to create and train a dense neural network and a convolutional neural network. Along with the data, there's a Word document, a Jupyter Notebook and a .py file which you can use to walk through the example.
Develop alternate models from the template
Once you've played with the example, you can select one of the data sets provided in the course classroom (CIFAR-10 or SVHN) on which to build your own deep learning network. Note that for each dataset, you can download a program that imports the data and structures it ready to go into a neural network. All you need to do is build and test your network.
Attachment:- Deep Learning.rar