[SOLVED] Cse 535 mobile computing project 2

The idea of this project is to use the Android Application created in Assignment 1 to upload
handwritten digit pictures to the server and use the server to classify the digits. The digits should
be placed in their respective folders after the classification of the images.
Project 2 will be a team project. Everyone is expected to submit a video of the application
working along with the source code.
Deliverables:
1) Mobile Application: The mobile application will be exactly similar to Assignment 1
except that there will not be any category dropdown list for this one.
2) Server Side: This time you will need to train a basic deep-learning framework from
scratch on the MNIST dataset to classify different handwritten digits. Once you have
trained the deep learning network on the dataset you will use the trained model to
classify the images as they are uploaded from the application. The classified images will
then need to be stored in their respective folder on the server. Please download the
dataset to train your model. For grading purposes, we will be using randomly clicked
images of handwritten digits. Your grades will be proportional to the accuracy of the
model.
TASKS: 1. Modify the Mobile Application 2. Modify the server code to accept the picture without
the image category 3. Decide on a ML Pipeline 4.Identify the dimensions of image 5.Identify the
type of Deep Learning framework that needs to be used 6.Build the Deep Learning Model from
scratch 7.Load the MNIST dataset 8.Split the dataset to train and validation 9.Preprocess the
dataset 10.Train the model using the dataset 11.Validate the trained model 12.Check the
weights of the Model and perform fine tuning if required 13.Store the trained model 14.Load the
trained model for testing 15.Check the accuracy of the model 16.Modify server to the call the
test function 17.Integrate all the components of the server 18.Check if the whole application is
integrated seamlessly 19.Provide some random handwritten digits and see the classification
20.Report the accuracy of the classification. 21.Check if the classified images are being stored
in the respective folders 22.Write the working of the application in the report format and create a
video of the working of the application 23.Put all the code and report in pdf format in a zip file
24. Submit the zipped file 25.Prepare the application for the seamless working demo
Submission:
1) Source Code of both Application and server.
2) Video of a working demonstration of the application and the server.
3) A 1-2 page report explaining the technical workings of your application.
All the above things need to be zipped together and uploaded on Canvas.

Important Dates for Project 1:

Notes:
We will be taking a Zero Tolerance Policy toward Plagiarism. So, please submit only your
original work. Violations of the University’s Academic Integrity policy will not be ignored.
Penalties include reduced or no credit for submitted work, a failing grade in the class, a note on
your official transcript that shows you were punished for cheating, suspension, expulsion, and
revocation of already awarded degrees. The university’s academic integrity policy can be found
at https://provost.asu.edu/academic-integrity.
Thank You