Write some software -- 2

I need you to develop some software for me. I would like this software to be developed . DUE FRIDAY MAY 26TH, 2017 AT 6:30PM EASTERN STANDARD TIME (EST). YOU NEED TO DO ONLY PROJECT PART 2! THE REST YOU DO NT NEED TO DO! Coding Project Assignment Project description: Using tensorflow and Keras to do the object recognition based on the MNIST data set. About the MNIST data set The MNIST database of handwritten digits, available from this page, has a training set of 60,000 examples, and a test set of 10,000 examples. It is a subset of a larger set available from NIST. The digits have been size-normalized and centered in a fixed-size image. More introduction of this data set is given below: [login to view URL] ###################################################################### Project part 1: Using single layer deep learning network to do the object recognition(getting the accuracy ) Project part 2: Using multi-layer deep learning network to do the object recognition Project part 3: Run the CNN large code to do the object recognition. Get the result and compare the execution time of the CNN large code to your single layer deep learning network code. Extra credit part: Option 1: Using single layer deep learning or multi-layer deep learning or CNN network to do something based on a new data set. Option 2: CNN large code of MNIST data set can be run based on the CPU version of tensorflow or the GPU version tensorflow. For the GPU version, it is not easy to be done and the main reason maybe we cannot easily find such a machine which contains the required machine such as the nvidia GPU. Meanwhile, if you have the GPU version, you can try to do the same experiment based on the GPU version. And then show me the execution time of the GPU version and then you can get the extra credit. For which GPU you can use to run the deep learning code, you can read this link to get more info. [login to view URL] ###################################################################### Team work: This project can be done by the cooperation of at most 3 people. Each people within the team will get the same grade. Submission requirement: Please submit the code, the result of your code(if the code is run from the terminal, pls send me the Screenshots of the result. If you run the code from the ipython notebook, please provide the result in your submission of the code) and a brief read me file to explain your experiment and result. Each team only need submit one copy of your work. Pls include all your team member’s name on your submission. Please submit it on the black board by one of your team members. Software requirement: python3.5 package installation: please install: numpy,tensorflow,keras, Tutorial about how to use Keras: [login to view URL]