Task 2. • Machine Leaming

Performdata pre-processing if it’s required.

1.

Formulate the problem as a rmchine leaming task.

2.

Select one leanmng algolithms based on the lectures and identify dae conesponding

3.

hyperparameters to be optimis ed.

Use the same data pa Ilitioning as the previous task.

4.

Perform model development

5.

o Choose an apptopriate feat-IR engineering method for in-age classification task and

extract your features. You need to investigate popular feature extraction methods for

image classification task and apply one of them

o Tram your rmdel based on training data and hyperparametertuning.

Perform performance assessment

6.

o Apply your NIL Imdel on the test data to get the prediction.

o Calculate the accuracy and the confusion nutlix.

(optional) Conduct other analvsis to be decided by the team nEmbers. For example:

7.

o Identify the best feat-um extraction method by con-paring at least 2 of popular ones.

2