代写COMP3032 – Machne Learning Assignment One代写C/C++语言
- 首页 >> Java编程COMP3032 - Machne Learning Assignment One (20 marks)
Marking guide
Task 1 (12 Marks):
Polynomial Regression Models (5 marks)
1 Read Data and Create Models (2 Marks)
o Properly read the pressure.csv dataset and assign the correct columns to the respective variables.
o Create polynomial regression models.
2 Plot (2 Marks)
o Produce across-validation error plot using the mean RMSE for degrees 1 to 14.
3 Print Coefficients (1 Mark)
o For the best degree of your choice, print the corresponding coefficients.
Multiple Linear Regression Model (2 marks)
4. Read Data and Create Model (2 marks)
o Read the dataset and create a multiple linear regression model to predict systolic pressure using all relevant features.
o Print the coefficients of the model.
Ridge Regression Model (2 marks)
5. Ridge Regression Model (2 marks)
o Build the ridge regression model and make predictions
o Print its coefficients.
Cross Validation (1 mark)
6. Cross-Validation (1 mark)
o Perform 10-fold cross-validation for the different models and display the mean RMSEs.
Short Answer and Documentation (2 marks)
7. Short Answer and Documentation
o Select the best degree and briefly explain why.
o Select the best model among the three and briefly explain why.
o Provide other necessary documentation (e.g., instructions on how to run the code, test run outputs, comments, etc.).
Task 2 (8 Marks): MNIST Digit Classification
1. Read Data (1 Mark)
o Read themnist_784 dataset and correctly assign the columns to the respective variables.
o Transform. the labels according to the requirements.
2. PCA (2 Marks)
o Perform Principal Component Analysis (PCA) on the feature data to reduce its dimensionality while retaining 90% of the overall explained variance ratio.
o Display the number of principal components preserved.
3. Logistic Model and Predictions (2 Marks)
o Create a logistic regression model using the reduced dataset.
o Properly split the dataset and use the model to predict the labels for both the training and testing sets.
4. Accuracy Evaluation (2 Marks)
o Calculate and display the accuracy, confusion matrix, and misclassified digits.
5. Short Answer and Documentation (1 Mark)
o Assess the model generated (e.g., good, underfit, overfit) and briefly explain why.
o Provide additional documentation (e.g., instructions on how to run the code, test run outputs, comments, etc.).