代写Assignment for Python Programming and Machine Learning代做留学生Python程序

Assignment for Python Programming and Machine Learning

Task 1: (50%)

Title: Application of Machine Learning in the Marine Industry: A Sector-Specific Review

Question:

Machine learning (ML) and artificial intelligence (AI) are transforming various industries, including the marine sector, by enhancing efficiency, optimizing operations, and improving safety. As students enrolled in different maritime disciplines—Shipping and Logistics, Naval Architecture, Marine Engineering, and Offshore Renewable Energy—you are required to conduct a critical review of how ML is applied in your respective subject area.

In your essay, you should:

1. Explain the fundamentals of machine learning and its relevance to the maritime industry.

2. Identify and discuss key ML applications within your chosen sector. Provide real-world examples where possible.

3. Analyse the benefits and challenges of ML implementation in your field, considering technical, economic, and environmental factors.

4. Evaluate future trends and potential advancements in ML that could impact your sector in the coming years.

5. Use appropriate references from academic papers, industry reports, and case studies to support your discussion.

Your essay should be well-structured, clearly written, and approximately 1,500 words in length.

Assessment Criteria:

· Depth and clarity of ML concepts (20%)

· Relevance and quality of examples (20%)

· Critical analysis of benefits and challenges (20%)

· Discussion of future trends (20%)

· Quality of writing and referencing (20%)

Task 2: (50%)

Artificial Neural Network (ANN) Model for Predicting Ultimate Strength of Unstiffened Plates Under Uni-Axial In-Plane Compression

Machine learning, particularly Artificial Neural Networks (ANNs), is widely used for predictive modelling in engineering applications. In this task, you are required to develop an ANN model to predict the ultimate strength of unstiffened plates under uni-axial in-plane compression based on provided dataset.

Your ANN model should follow these specifications:

· Architecture: One input layer, one hidden layer, and one output layer.

· Input Variables:

1. Plate width

2. Plate thickness

3. Initial imperfection (residual stress and deflection due to welding)

4. Yield stress of the material

· Output Variable: Ultimate strength of the plate.

Steps to Follow:

1. Preprocess the data:

o Load the dataset from the provided spreadsheet.

o Handle missing values, normalize the data, and split into training and testing sets.

2. Build the ANN model:

o Choose an appropriate activation function for each layer.

o Train the model using a suitable optimizer and loss function.

o Tune hyperparameters such as the number of neurons in the hidden layer and the learning rate.

3. Evaluate the model:

o Use appropriate performance metrics (e.g., Mean Squared Error, R² score).

o Perform. validation using unseen test data.

4. Interpret results and discuss limitations:

o Analyze the performance of your ANN model.

o Discuss how well it predicts the ultimate strength of unstiffened plates.

o Identify potential improvements or challenges in applying ANN models in real-world scenarios.

Alternative Case Study Selection:

Students could possibly choose a case relevant to their subject area (Shipping and Logistics, Naval Architecture, Marine Engineering, Offshore Renewable Energy). The selected case study must be discussed with the lecturer before implementation.

Deliverables:

· Python script. (Jupyter Notebook or .py file) implementing the ANN model.

· A short report (< 500 words) describing the model development, data processing, results, and discussion.

· Graphs and tables illustrating model performance and comparisons.

Assessment Criteria:

· Data preprocessing and handling (20%)

· ANN model development and implementation (40%)

· Evaluation and interpretation of results (20%)

· Quality of report (20%)

Submission Format:

· The Python code should be submitted as a Jupyter Notebook (.ipynb).

· The report should be submitted as a WORD document.

· All necessary data and results should be clearly documented.