Explore

F23 AML3304 Topics Syllabus for the Mid Term Test

The Test will be 90 minutes in duration:

Delivered by ClassMarker URL in the last part of class.

50 Questions:

40 MCQ

10 Short Answer

As a way of studying, you can use CHATGPT as your Training Coach. See this article for details:

⁠

https://www.linkedin.com/pulse/chatgpt-prompt-use-do-practice-interview-sessions-peter-sigurdson⁠

⁠

Learning Resources:

Concepts Map:

https://www.edrawmax.com/online/share.html?code=e4dbc64471b911eeb4cd0a951ba8b83d⁠

⁠

Mind Map:

⁠

https://www.edrawmax.com/online/share.html?code=e4dbc64471b911eeb4cd0a951ba8b83d⁠

⁠

PowerPOINT:

⁠

https://www.dropbox.com/scl/fi/ljcsyy2qazu1v1ibiw9y5/AML3304-F23.pptx?rlkey=2xyq2xep1nuxqehr51ftueqnc&dl=0⁠

⁠

Moodle Day Activities Journal:

https://moodle.cestarcollege.com/moodle/mod/page/view.php?id=1467716⁠

⁠

Midterm will be delivered with Classmarker:

15 questions

Every question type: TF, MCQ, Match up the concepts, short answer.

Chain of reasoning type questions, in which you will be given a story, and asked to deduce or aduce facts or evaluations based on that story.

You may be asked to write some Python code based on lab activities.

You may be shown code: and asked what it will output.

You might get some “red herring” questions.

Describe the Software Crisis. What were its structural components, why did it happen? How did the Software Crisis inform our thinking today as to how we should do Software Project Management and Software Engineering?

Questions and Topics to study:

Use the PYTHON Library PYTORCH to output the PYTORCH Tensor File:

⁠

https://coda.io/@peter-sigurdson/introduction-to-pytorch-as-an-essential-ml-framework⁠

Describe the Layers of the AI MODEL Architecture? What is the importance of each layer?

⁠

https://coda.io/@peter-sigurdson/lecture-architecture-of-the-ai-application⁠

⁠

Why is JSON and Big Data a topic in building deploying the ai model?

PFEQ: Why can’t we use SQL data store for the AI MODEL:

Because SQL is based on the Primary Key: Codd’s Laws wants to wash out the subtle details of the Data Interactions. (The purpose of the Primary Key is to wash out the subtle interactions of the data. Normalize means “Make everything look the same”. Great for highly structured data.

When we apply Baysian Training, to a Training Corpus of Input Data, the PYTORCH PYTHON library will use methods to output an Edged Graph Data Structure which encodes the TOKENS {the words in the INPUT training Corpus} and the connections between those words {which is the Weightings} into a Matrix. This output Matrix is the PYTORCH tensor file which you upload to a Server for Users to connect to and hold conversations with.

JSON is the Data Format that makes this conversational memory possible. WHY?

⁠

Want to print your doc?
This is not the way.

Try clicking the ⋯ next to your doc name or using a keyboard shortcut (

CtrlP

) instead.