DTS101TC代做、代写Python语言程序

项目预算：开发周期：发布时间：要求地区：

DTS101TC Coursework
This coursework is designed to assess your understanding of neural networks and machine learning concepts, as well as your ability to implement, analyze, and evaluate models effectively. It consists of two main components: five assignments and an image object detection project. Detailed instructions, marking criteria, and submission requirements are outlined below. AIGC tools are not allowed.

Part 1: Assignments (50 Marks)
This section includes five individual assignments, each focusing on different neural network techniques and datasets. The breakdown for each task includes marks for code execution, analysis, evaluation, and reporting quality.
Submission Requirements
Please submit your notebooks to Gradescope. Each assignment must be completed according to the instructions provided in the Python Jupyter Notebook, with all output cells saved alongside the code. You don’t need to write a report for this part. Please put all the analysis and results in your notebook.
Weekly TA checks during lab sessions and office hours are mandatory. Assignments will not be graded without TA verification.
Question 1: Digit Recognition with Neural Networks
Task: Implement a basic neural network using TensorFlow/PyTorch to train a digit recognition model on the MNIST dataset.
Mark Breakdown:
oCode execution by Gradescope: 5 marks
oData and model analysis: 2 marks
oTest cases: 2 marks
oReport quality (comments and formatting): 1 mark
Question 2: Logistic Regression for Flower Classification
Task: Build and implement a Logistic Regression model to classify three types of iris flowers using the dataset in sklearn.
Mark Breakdown:
oCode execution by Gradescope: 5 marks
oData and model analysis: 2 marks
oTest cases: 2 marks
oReport quality (comments and formatting): 1 mark

Question 3: House Price Prediction with ANN/MLP
Task: Design and implement an ANN/MLP model to predict house prices in California using the dataset in sklearn.
Mark Breakdown:
oCode execution by Gradescope: 5 marks
oData and model analysis: 2 marks
oTest cases: 2 marks
oReport quality (comments and formatting): 1 mark
Question 4: Stock Price Prediction with RNN
Task: Create an RNN model to predict stock prices for companies like Apple and Amazon from the Nasdaq market using the provided dataset.
Mark Breakdown:
oCode execution by Gradescope: 5 marks
oData and model analysis: 2 marks
oModel evaluation: 2 marks
oReport quality (comments and formatting): 1 mark
Question 5: Image Classification with CNN
Task: Develop a CNN model to classify images into 10 classes using the CIFAR-10 dataset.
Mark Breakdown:
oCode execution by Gradescope: 5 marks
oData and model analysis: 2 marks
oModel evaluation: 2 marks
oReport quality (comments and formatting): 1 mark

Part 2: Project (50 Marks)
The project involves building a custom image dataset and implementing an object detection neural network. This is a comprehensive task that evaluates multiple skills, from data preparation to model evaluation.
Submission Requirements
All of your dataset, code (Python files and ipynb files) should be a package in a single ZIP file, with a PDF of your report (notebook with output cells, analysis, and answers). INCLUDE your dataset in the zip file.
Step 1: Dataset Creation (10 Marks)
Task: Collect images and use tools like Label Studio or LabelMe to create labeled datasets for object detection. You can add one more class into the provided dataset. The dataset should have up to 10 classes. Each contains at least 200 images.
Deliverable: Include the dataset in the ZIP file submission.
Mark Breakdown:
oCorrect images and labels: 6 marks
oData collection and labeling process explanation: 2 marks
oDataset information summary: 2 marks
Step 2: Data Loading and Exploration (10 Marks)
Task: Organize data into train, validation, and test sets. Display dataset statistics, such as class distributions, image shapes, and random samples with labels. Randomly plot 5 images in the training set with their corresponding labels.
Mark Breakdown:
oCorrect dataset splitting: 6 marks
oDataset statistics: 2 marks
oSample images and labels visualization: 2 marks
Step 3: Model Implementation (10 Marks)
Task: Implement an object detection model, such as YOLOv8. Include a calculation of the total number of parameters in your model. You must include calculation details.
Mark Breakdown:
oCode and comments: 6 marks
oParameter calculation details and result: 4 marks
Step 4: Model Training (10 Marks)
Task: Train the model using appropriate hyperparameters (e.g., epoch number, optimizer, learning rate). Visualize training and validation performance through graphs of loss and accuracy.
Mark Breakdown:
oCode and comments: 6 marks
oHyperparameters analysis: 2 marks
oPerformance analysis: 2 marks
Step 5: Model Evaluation and Testing (10 Marks)
Task: Evaluate the model on the test set, displaying predictions (visual result) and calculating metrics like mean Average Precision (mAP) and a confusion matrix.
Mark Breakdown:
oCode and comments: 6 marks
oPrediction results: 2 marks
oEvaluation metrics: 2 marks
Submission Guidelines
1.Assignments: Submit your Jupyter Notebooks via Gradescope. Ensure all output cells are saved and visible.
2.Project: Submit your ZIP file containing the dataset, Python files, Jupyter Notebooks, and a PDF report via Learning Mall Core.
General Notes and Policies
1.Plagiarism: Submissions must be your own work. Avoid copying from external sources without proper attribution. Sharing code is prohibited.
2.Late Submissions: Follow the university's policy on late submissions; penalties may apply.
3.Support: Utilize lab sessions and TA office hours for guidance.

Marking Criteria
Assignments
Code execution by Gradescope: 5 marks
Data and model analysis: 2 marks
Test cases or model evaluation: 2 marks
Report quality (comments and formatting): 1 mark
Project
Code (60%):
oFully functional code with clear layout and comments: 6 marks
oPartially functional code with some outputs: 4 marks
oCode that partially implements the solution but does not produce outcomes: 2 marks
oIncomplete or non-functional code: 0 marks
Analysis (40%):
oComplete and accurate answers with clear understanding: 4 marks
oPartial answers showing some understanding: 2 marks
oLimited understanding or incorrect answers:: 0 marks