Insert Here Your Project Title

Submitted by: InsertTeamNameHere

Submitted to: Hao Lac

Discipline: Software Engineering

Due Date: InsertDueDateHere

1
Declaration of Sole Authorship
We, insertTeamNameHere, confirm that this work submitted for assessment is

our own and is expressed in our own words. Any uses made within it of the works

of any other author, in any form (ideas, equations, figures, texts, tables,

programs), are properly acknowledged at the point of use. A list of the references

used is included.

Signed: insertMemberSignatureHere, insertStudentIDHere (programOfStudy)

Date: insertDueDateHere

2
 Abstract

Provide an abstract in this section in 200 words or less in paragraph format.

An accurate summary of the TR. State the main idea or thesis by answering

questions such as:

● What is the TR about?

● Why is it significant?

● What should I do about it?

3
Table of Contents
Declaration of Sole Authorship
Abstract
List of Figures
1.0 INTRODUCTION
2.0 METHODOLOGY AND RESULTS
2.1 Literature Review
2.2 Proposed Solution
2.3 Users of the Proposed System
2.4 DataFrame Step
2.5 Transformer Step
2.6 Estimator Step: the Learning Algorithm & Cost Function
2.7 Estimator Step: Hyperparameters and Fine Tuning
2.8 Evaluator Step: Training and Testing
3.0 CONCLUSIONS
4.0 FUTURE WORKS
CREDITS, LICENSE, AND REFERENCES
Credits
License
References

4
List of Figures

Figure 1: Modification of Machine Learning Pipeline from Google Image. 10

5
1.0 INTRODUCTION

Introduction (Including the problem statement):

● What is the technical problem?

● Why was the work described in the TR undertaken?

● What is included and/or omitted? What is the scope of the report and what

procedures are used?

● What is your objective?

● What unique problems were encountered in doing or interpreting the

work?

● Are there unique approaches in the study?

6
2.0 METHODOLOGY AND RESULTS

2.1 Literature Review

Provide a literature review of existing solutions to the problem discussed in the

previous section. Focus your discussion on the strengths and weaknesses of

these existing solutions.

2.2 Proposed Solution

Provide a description of your proposed solution. Discuss the strengths and

weaknesses of your system.

Provide a developer perspective diagram of your software/hardware network

architecture with proper captioning and accompanying explanations in the body

of the text. A sample diagram can be found here.

7
2.3 Users of the Proposed System

Discuss the users of your system as this applies to how the users interface with

your system.

For each users of your system, include detail that answer at least the following

questions:

● The frequency with which the user will use the software.

● The user's level of expertise with the domain.

● The user's general level of proficiency with computers and software.

● The user's level of proficiency with the software being developed.

● The user's general goal for using the software. Some users are after

convenience, others favour a rich experience, and so on.

8
2.4 DataFrame Step

Provide a discussion of the dataset used in your project with citation of the

original source (e.g.,

https://archive.ics.uci.edu/dataset/80/optical+recognition+of+handwritten+digits).

2.5 Transformer Step

Provide a detailed discussion of the following aspects of the Transformer step

(refer to Figure 1) with respect to your project:

● The feature extraction approach used on your dataset; and

● The data cleaning approach used on your dataset.

In some applications such as text analysis, feature extraction can be applied to

the raw data twice; once before data cleaning and again after data cleaning

because some features may be lost if cleaning precedes feature extraction such

as uppercase words.

For other problems such as image recognition, image processing such as image

restoration or Independent Component Analysis are usually applied to images to

extract information from it or enhance it. In fact, the task of cleaning and

enhancing images can be an entire project unto itself by training an ANN to

perform these tasks.

All in all, the step prior to training your system is to transform your data to

enhance the system’s performance capability.

9
If for whatever reason your project does not apply an explicit feature extraction

step and/or a data cleaning step (extremely rare), you will need to provide a

detailed, quantitative analysis of how this negatively impacts your training time

and most likely, your system’s performance.

2.6 Estimator Step: the Learning Algorithm & Cost Function

All ML systems have an algorithm of how to update their weights, which are the

constant value of linear combinations of vectors. In this section, provide the

following with respect to the goal of maximizing your system’s generalization

capability:

● the learning algorithm used; and

10
 ● the Cost Function optimized.

2.7 Estimator Step: Hyperparameters and Fine Tuning

All non-trivial ML systems have hyperparameters who’s value have to be

tweaked for optimal results. In this section, provide a detailed discussion of the

following with respect to the goal of maximizing your system’s generalization

capability:

● The hyperparameters in your system (e.g., number of nodes, number

layers, number of epochs, etc.); and

● The approach used to train the dataset to your system to maximize its

generalization capability (i.e., # of epochs, # of iterations, batch,

incremental, online, cross-validation, hybrid methods, etc.).

2.8 Evaluator Step: Training and Testing

Discuss the generalization capability of your system as follows:

● Show visually (i.e., graph) the distribution of your validation and test

datasets so that when you feed your test dataset into your trained system,

you can ensure that the results are not biased; and

● Provide a quantitative analysis (statistical confidence) of your system’s

performance on the test dataset.

3.0 CONCLUSIONS

Provide a conclusion that interprets the data and results of your project. It is

11
reasoned judgment and not opinions. Consider the variables. Related cause and

effect. Analyze, evaluate, make comparisons and contrasts. Base the conclusion

on fact.

4.0 FUTURE WORKS

Provide a discussion of future works presuming you will be continuing to work

and improve upon your project.

12
CREDITS, LICENSE, AND REFERENCES

Credits

Provide any credits here. The following are examples:

Author of the template graphic layout : Hao Lac

<[email protected]>

Author of the template explanation text : John Doe

<[email protected]>

License

State the license granted with your system. For example:

Permission is granted to copy, distribute and/or modify this document under the

terms of the GNU Free Documentation License, Version 1.1 or any later version

published by the Free Software Foundation; with no Invariant Sections, with no

Front-Cover Texts, and with no Back-Cover Texts. A copy of the license is

included in the appendix entitled "GNU Free Documentation License".

References

[1] TAC: Technology Accreditation Canada,

https://www.technologyaccreditation.ca/.

13