andr3/fyp-report
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases 3 Wiki Activity

update project synopsis
All checks were successful
continuous-integration/drone/push Build is passing
Details

Browse Source
This commit is contained in:
Andre Henriques 2023-11-02 13:11:40 +00:00
parent a683485b70
commit 90d4b0e4bb
1 changed files with 23 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
projectsynopsis/project-synopsis.tex
View File

@ -57,17 +57,12 @@
\section{Introduction} \section{Introduction}
% This section should contain an introduction to the problem aims and objectives (0.5 page) % This section should contain an introduction to the problem aims and objectives (0.5 page)
The aim of this project is to create a classification service that has 0 requires zero user knowledge about machine learning, image classification or data analysis.
The aim of this project is to create a classification service that has 0 The system should allow the user to create a reasonable accurate model that can satisfy the users' need.
requires zero user knowledge about machine learning, image The system should also allow the user to create expandable models; models where classes can be added after the model has been created.
classification or data analysis.
The system should allow the user to create a reasonable accurate model
that can satisfy the users' need.
The system should also allow the user to create expandable models;
models where classes can be added after the model has been created.
\subsection{Aims} \subsection{Aims}
The project aims to create a platform where users can create different types of classification models without the users having any knowledge of image classification. The project aims to create a platform where users can create different types of classification models without the users having any knowledge of image classification.
\subsection{Objectives} \subsection{Objectives}
This project's primary objectives are to: This project's primary objectives are to:
@ -86,33 +81,40 @@
\section{Literature and Techincal Review} \section{Literature and Techincal Review}
% 1 page of background and literature review. Here you will need to references things. Gamal et al.~\cite{gamal} introduce the concept of \ldots % 1 page of background and literature review. Here you will need to references things. Gamal et al.~\cite{gamal} introduce the concept of \ldots
\subsection{Alternatives to my Project} \subsection{Alternatives to my Project}
There currently exist systems that do image classification, like Google Vision AI\cite{google-vision-api}, and Amazon's Rekoginition\cite{amazon-rekognition}. There currently exist systems that do image classification, like Google Vision AI\cite{google-vision-api}, and Amazon's Rekoginition\cite{amazon-rekognition}.
Their tools, while providing similar services to what my project is supposed to do, it mostly focusses on general image classification rather than specific image classification, i.e. Car vs Boat, vs, Car model X vs Car model Y. Their tools, while providing similar services to what my project is supposed to do, it mostly focusses on general image classification rather than specific image classification, i.e. Car vs Boat, vs, Car model X vs Car model Y.
\subsection{Creation Models} \subsection{Creation Models}
The models that I will be creating will be Convolutional Neural Network(CNN)\cite{lecun1989handwritten,fukushima1980neocognitron}. The models that I will be creating will be Convolutional Neural Network(CNN)\cite{lecun1989handwritten,fukushima1980neocognitron}.
The system will be creating two types of models that cannot be expanded and models that can be expanded. For the models that can be expanded, see the section about expandable models. The system will be creating two types of models that cannot be expanded and models that can be expanded. For the models that can be expanded, see the section about expandable models.
The models that cannot be expanded will use a simple convolution blocks, with a similar structure as the AlexNet\cite{krizhevsky2012imagenet} ones, as the basis for the model. The size of the model will be controlled by the size of the input image, where bigger images will generate more deep and complex models. The models that cannot be expanded will use a simple convolution blocks, with a similar structure as the AlexNet\cite{krizhevsky2012imagenet} ones, as the basis for the model. The size of the model will be controlled by the size of the input image, where bigger images will generate more deep and complex models.
The models will be created using TensorFlow\cite{tensorflow2015-whitepaper} and Keras\cite{chollet2015keras}. These theologies are chosen since they are both robust and used in industry. The models will be created using TensorFlow\cite{tensorflow2015-whitepaper} and Keras\cite{chollet2015keras}. These theologies are chosen since they are both robust and used in industry.
\subsection{Expandable Models} \subsection{Expandable Models}
The current most used approach for expanding a CNN model is to retrain the model. This is done by, recreating an entire new model that does the new task, using the older model as a base for the new model, or using a pretrained model as a base and training the last few layers. The current most used approach for expanding a CNN model is to retrain the model. This is done by, recreating an entire new model that does the new task, using the older model as a base for the new model\cite{amazon-rekognition}, or using a pretrained model as a base and training the last few layers.
There are also unsupervised learning methods that do not have a fixed number of classes. While this method would work as an expandable model method, it would not work for the purpose of this project. This project requires that the model has a specific set of labels which does not work with unsupervised learning which has unlabelled data. Some technics that are used for unsupervised learning might be useful in the process of creating expandable models.
\section{Technical overview} \section{Technical overview}
\subsection{Interface} % 1 page of overview. My approach is shown in Figure~\ref{fig:sample}. You can draw the diagram in powerpoint and save the picture
\subsection{Web Interface}
The user will interact with the platform form via a web portal. The user will interact with the platform form via a web portal.
The web platform will be designed using HTML and a JavaScript library called HTMX\cite{htmx} for the reactivity that the pagers requires. The web platform will be designed using HTML and a JavaScript library called HTMX\cite{htmx} for the reactivity that the pagers requires.
The web server that will act as controller will be implemented using go\cite{go}, due to its ease of use. The web server that will act as controller will be implemented using go\cite{go}, due to its ease of use.
The web server will also interact with python to create models. Then to run the models, it will use the libraries that are available to run TensorFlow\cite{tensorflow2015-whitepaper} models for that in go. The web server will also interact with python to create models. Then to run the models, it will use the libraries that are available to run TensorFlow\cite{tensorflow2015-whitepaper} models for that in go.
\subsection{Creating Models}
The models will be created using TensorFlow.
The original plan was to use go and TensorFlow, but the go library was lacking that ability. Therefore, I chose to use python to create the models.
The go server starts a new process, running python, that creates and trains the TensorFlow model. Once the training is done, the model is saved to disk which then can be loaded by the go TensorFlow library.
\subsection{Expandable Models} \subsection{Expandable Models}
The approach would be based on multiple models. The first model is a large model that will work as a feature traction model, the results of this model are then given to other smaller models. These model's purpose is to classify the results of the feature extraction model into classes. The approach would be based on multiple models. The first model is a large model that will work as a feature traction model, the results of this model are then given to other smaller models. These model's purpose is to classify the results of the feature extraction model into classes.
The first model would either be an already existent pretrained model or a model that is automatically created by the platform. The first model would either be an already existent pretrained model or a model that is automatically created by the platform.
The smaller models would all be all generated by the platform, this model's purpose would be actually classification. The smaller models would all be all generated by the platform, this model's purpose would be actually classification.
This approach would offer a lot of expandability, as it makes the addition of a new class as easy as creating a new small model. This approach would offer a lot of expandability, as it makes the addition of a new class as easy as creating a new small model.
% 1 page of overview. My approach is shown in Figure~\ref{fig:sample}. You can draw the diagram in powerpoint and save the picture
\section{Workplan} \section{Workplan}
\subsection{Timeline} \subsection{Timeline}