You built a model which classifies if you're a python-wiz or a python-was. Calculate the F1-Score based on the following data: Correctly identified as a python-wiz: 126 Correctly identified as a python-was: 58 Incorrectly identified as a python-wiz: 14 Incorrectly identified as a python-was: 22 0.85 0.89 0.88 0.84 You built a model to detect COVID-19 cases in rural areas based on daily population movements. Your classifier had the following results: Correctly identified negative cases: 100241 Correctly identified positive cases: 95 Incorrectly identified negative cases: 24020 Incorrectly identified positive cases: 28 We need to measure the model's performance in identifying individuals needing to quarantine. False positives will lose 2 weeks of pay in quarantine whilst being healthy. False negatives will continue going to work, and infecting others. With this understanding, plus the fact that the data has a class imbalance. Which of the following metrics would best assess your model's performance? F1-Score Accuracy Precision Recall When building a logistic model we can do the following: Change the value of C to -1.00 Automatically optimise the C parameter in the LogisticRegression() object Change the probability threshold from 0.5 to 0.6 Assign a penalty of 12 and 11 Assuming you have done the following: 1) imported LogisticRegression from sklearn.linear_model 2) imported train_test_split from sklearn.model_selection 3) cleaned and split your dataset to into 'X' and 'y Which lines of code have an error? = i) X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.25, random_state=42) ii) model = LogisticRegression(C=0.1, penalty='11') iii) model.fit(X_train_y_train) iv) pred = model.predict(y_test) iv) iii), iv) i), iv) ii) Which of the following statements is CORRECT? The more data we have, the faster our logistic regression model trains. The logistic regresssion model can detect and remove outliers in the data. O Logistic Regression is a parametric model whereas a K-Nearest Neighbours model is non-parametric. We can get more accurate predictions on a binary classification dataset by training a logistic regression model on each class individually and then adding the two model predictions together