Deep Learning and Convolutional Neural Networks | My Assignment Tutor

42028: Assignment 2 – Autumn 2021 Page 1 of 4 Faculty of Engineering and Information TechnologySchool of Software 42028: Deep Learning and Convolutional Neural NetworksAutumn 2021ASSIGNMENT-2 SPECIFICATION Due dateFriday 11:59pm, 14 May 2021DemonstrationsOptional, If required.Marks40% of the total marks for this subjectSubmission1. A report in PDF or MS Word document (~10-pages)(Part-B submission)2. Google Colab/iPython notebooks (Part-A submission)Submit toCanvas assignment submissionNote: This assignment is an individual work. Your assignment isincomplete without the Report and Code submission. If you just submiteither code or the report, it will considered incomplete and will not bemarked. Please make sure to submit both the report and code using theappropriate links/pages on Canvas. SummaryThis assessment requires you to customize the standard CNN architectures forimage classification. Standard CNNs such as AlexNet, GoogleNet, ResNet, etc.should be used to create customized version of the architectures. Students arealso required to implement a custom CNN architecture for object detection andlocalization. Both the customized CNNs (image classification and object detection)should be trained and tested using the dataset provided.Students need to provide the code (ipython/Colab Notebook) and a final reportfor the assignment, which will outline a brief assumptions/intuitions considered tocreate the customized CNNs and discuss the performance.Assignment ObjectivesThe purpose of this assignment is to demonstrate competence in the followingskills. To ensure that the student has a firm understanding of CNNs and objectdetections algorithms. This will facilitate the learning of advanced topics forresearch and also assist in completing the project.42028: Assignment 2 – Autumn 2021 Page 2 of 4 To ensure that the student can develop custom CNN architectures for differentcomputer vision related tasks.Tasks:Description:1. Customize AlexNet/GoogleNet/ResNet etc. and reduce/increase the layers,Train and test for image classification.2. Implement the Faster-RCNN and SDD architectures for objectdetection/localization. (Use of existing implementation such as Google Objectdetection API is permitted).3. Train and test on the given dataset for object detection, using Faster-RCNN andSSD object detection methods.Datasets for each tasks will be provided.Write a short report on the implementation, linking the concepts and methodslearned in class, and also provide assumptions/intuitions considered to create thecustom CNNs for image classification. Provide diagrams for the CNNs architecturewhere required for better illustrations. Provide the model summary, such as inputand output parameters, etc. Discuss the results clearly and explain the differentsituations/constraints for the better understanding of the results obtained.Dataset to be used: Provided separately (Check Canvas underAssignmentAssignment-2).Report Structure (suggestion only):The report may include the following sections:1. Introduction: Provide a brief outline of the report and also briefly explainthe baseline CNN architectures used to create the custom CNNs for imageclassification. Also mention about the object detection methods used.2. Dataset: Provide a brief description of the dataset used with some sampleimages of each class.3. Proposed CNN architecture for Image classification:a. Baseline architecture used.b. Customized architecturec. Assumptions/intuitionsd. Model summary4. CNN architecture for Object Detection/localization:a. Faster-RCNN.b. SSD (Single Shot Detector)c. Assumptions/intuitionsd. Model summary5. Experimental results and discussion:a. Experimental settings:i. Image classificationii. Object detection42028: Assignment 2 – Autumn 2021 Page 3 of 4b. Experimental Results:i. Image classification1. Performance on baseline/standard architecture2. Performance on customized architectureii. Object detection1. Performance on Faster-RCNN2. Performance on SSD or customized architecture.iii. Discussion: Provide your understanding of the performanceand accuracy obtained. You may also include some imagesamples which were wrongly classified.6. Conclusion: Provide a short paragraph summarizing your understanding ofthe experiments and results.Deliverables:a. Project Report (10 pages max)b. Google Colab or Ipython notebook, with the codeNote: You are welcome to report accuracy on custom CNNs designed for Objectdetection, instead of SSD. Use of transfer learning permitted.Additional Information:Assessment SubmissionSubmission of your assignment is in two parts. You must upload the Ipython/Colabnotebooks (zip-file in case of multiple notebooks) and Report to Canvas. This must bedone by the Due Date. You may submit as many times as you like until the due date.The final submission you make is the one that will be marked. If you have not uploadedyour zip file within 7 days of the Due Date, or it cannot be run in the lab, then yourassignment will receive a zero mark. Additionally, the result achieved and shown in theipython/Colab notebooks should match the report. Penalties apply if there areinconsistencies in the experimental results and the report.PLEASE NOTE 1: It is your responsibility to make sure you have thoroughly tested yourprogram to make sure it is working correctly.PLEASE NOTE 2: Your final submission to Canvas is the one that is marked. It does notmatter if earlier submissions were working; they will be ignored. Download yoursubmission from Canvas and test it thoroughly in your assigned laboratory.Return of Assessed AssignmentIt is expected that marks will be made available 2 weeks after the submission via Canvas.You will be given a copy of the marking sheet showing a breakdown of the marks.42028: Assignment 2 – Autumn 2021 Page 4 of 4QueriesIf you have a problem such as illness which will affect your assignment submissioncontact the subject coordinator as soon as possible.Dr. Nabin SharmaRoom: CB11.07.124Phone: 9514 1835Email: [email protected] you have a question about the assignment, please post it to the Canvas forum forthis subject so that everyone can see the response.If serious problems are discovered the class will be informed via an announcement/FAQson Canvas. It is your responsibility to make sure you frequently check Canvas.PLEASE NOTE: If the answer to your questions can be found directly in any of thefollowing Subject outline Assignment specification Canvas FAQ Canvas discussion boardYou will be directed to these locations rather than given a direct answer.Extensions and Special ConsiderationPlease refer to subject outline.Academic Standards and Late PenaltiesPlease refer to subject outline.


Leave a Reply

Your email address will not be published. Required fields are marked *