This document outlines the requirements for Assignment $2.$ In this assignment, you are required to implement a Convolutional Neural Network $($CNN$)$ using Python's TensorFlow$/$Keras library. You may select your own or publicly available image data and explore a topic relevant to CNN applications.

Your findings and results must be summarised in a poster presentation, which should clearly and concisely communicate your project and its outcomes.

REQUIREMENTS

Your poster and accompanying submission must include the following:

Title and Introduction:

Clearly state the topic of your project and the problem you aim to solve using CNNs$.$

Model Architecture:

Provide a detailed description of your CNN model, including:

The number and type of layers $($e$.$g$.,$ convolutional, pooling, dense$).$

The types of activation functions used $($e$.$g$.,$ ReLU, softmax$).$

Regularisation Techniques:

Explain the methods you used to reduce overfitting $($e$.$g$.,$ dropout, L$2$ regularisation, data augmentation$).$

Optimisation Algorithm:

Specify the optimizer $($e$.$g$.,$ Adam, SGD$)$ $ind hyperparameters used $($e$.$g$.,$ learning rate, batch size$).$

Performance Evaluation:

Describe how you assessed the performance of your model, including metrics $($e$.$g$.,$ accuracy, F$1-$score$)$ and benchmarks for comparison.

Poster Presentation:

Summarise your project in a visually appealing poster that includes:

Introduction$/$Background: Briefly describe the context and significance of your work.

Model Design: Outline the CNN architecture and techniques used.

Results: Present key findings using charts, graphs, or images.

Conclusion: Highlight insights, challenges, and potential improvements.

Code and Dataset:

Submit the Python code used to develop the CNN$.$

Provide access to the dataset used $($or a link to it$,$ if publicly available$).$This document outlines the requirements for Assignment $2.$ In this assignment, you are required to implement a Convolutional Neural Network $($CNN$)$ using Python's TensorFlow$/$Keras library. You may select your own or publicly available image data and explore a topic relevant to CNN applications.

Your findings and results must be summarised in a poster presentation, which should clearly and concisely communicate your project and its outcomes.

REQUIREMENTS

Your poster and accompanying submission must include the following:

Title and Introduction:

Clearly state the topic of your project and the problem you aim to solve using CNNs$.$

Model Architecture:

Provide a detailed description of your CNN model, including:

The number and type of layers $($e$.$g$.,$ convolutional, pooling, dense$).$

The types of activation functions used $($e$.$g$.,$ ReLU, softmax$).$

Regularisation Techniques:

Explain the methods you used to reduce overfitting $($e$.$g$.,$ dropout, L$2$ regularisation, data augmentation$).$

Optimisation Algorithm:

Specify the optimizer $($e$.$g$.,$ Adam, SGD$)$ $ind hyperparameters used $($e$.$g$.,$ learning rate, batch size$).$

Performance Evaluation:

Describe how you assessed the performance of your model, including metrics $($e$.$g$.,$ accuracy, F$1-$score$)$ and benchmarks for comparison.

Poster Presentation:

Summarise your project in a visually appealing poster that includes:

Introduction$/$Background: Briefly describe the context and significance of your work.

Model Design: Outline the CNN architecture and techniques used.

Results: Present key findings using charts, graphs, or images.

Conclusion: Highlight insights, challenges, and potential improvements.

Code and Dataset:

Submit the Python code used to develop the CNN$.$

Provide access to the dataset used $($or a link to it$,$ if publicly available$).$