$1.$ Use the dataset called "credit$\_$risk.csv$".$Use $\text{'}$CSV File Import' widget to upload the dataset.Use 'Select Columns' widget to set 'loan$\_$status' as the target variable$\mathrm{.}2.$ Use the Python Script called "create$\_$deterministic$\_$sample.py$"$ to create a random sample of the dataset.Go to the bottom of the Python Script and replace the default value of $42$ for the random$\_$seed with the numeric part of your UID $($you can drop leading zero$).$ Take a screenshot to show that you changed the random seed and save it as "Random Seed".Leave sample$\_$size as $0\mathrm{.}8\mathrm{.}3.$ Connect 'Select Columns' to this 'Python Script' and connect the output to a 'Test and Score' widget. In 'Test and Score' widget, select 'Cross validation' and 'Number of folds' as $5.$ This is your full model. Open the 'Select Columns' widget and take a screenshot and save it as "Full Model"$\mathrm{.}4.$ Bring the 'Neural Network' widget on to the workflow canvas and rename it as "Neural Network $($Config $1)".$ Before connecting it:Uncheck the 'Apply Automatically' checkbox.Change the maximal number of iterations to $10.$After making those changes, take a screenshot of the 'Neural Network' properties and save it as "Neural Network Config $1"\mathrm{.}5.$ Connect 'Neural Network' to 'Test and Score'. Make sure to go into the 'Neural Network' widget and click on 'Apply' $($remember$,$ you unchecked the 'Apply Automatically' in Step $4,$ so now you have to 'Apply' it manually$)\mathrm{.}6.$ Create two additional models using 'Select Columns' $($i$.$e$.,$ keep the same target variable, but select different independent variables to include$).$ Call them 'Model $1\text{'}$ and 'Model $2\text{'}.$ Take screenshots of the opened 'Select Columns' widgets for both models and save them as "Model $1"$ and "Model $2"\mathrm{.}7.$ Capture the evaluation results $($AUC$,$ CA$,$ F$1,$ Precision, Recall, and MCC$)$ for each model from the 'Test and Score' widget in an Evaluation Results Table $($as shown below$)\mathrm{.}8.$ Make a copy of the 'Neural Network' widget and rename it as "Neural Network $($Config $2)".$ Disconnect 'Neural Network $($Config $1)"$ from the three 'Test and Score' widgets and connect the 'Neural Network $($Config $2)$ to the three 'Test and Score' widgets.Now, change one or more the following properties of the 'Neural Network $($Config $2)\text{'}$ widget: 'Neurons in hidden layers', 'Maximal number of iterations', 'Activation' function, or 'Solver'. Modify the properties of 'Neural Network $($Config $2)\text{'}$ in such a way that it improves at least one of the evaluation metrics for the three models. The metric that improves can be different for the different models. Take a screenshot of the neural network properties and save it as "Neural Network Config $2".$Capture the evaluation results just as in Step $7$ in the Evaluation Results table $($as shown below$).$