Hands-On Data Analysis with Pandas

Presenting our results

The SOC has finished up the March data. They want us to implement into our model the feedback they gave on our February predictions, and then make predictions for the March data for them to review. They will be evaluating our performance on each minute in March, using the classification report, ROC curve, confusion matrix, and precision-recall curve. It's time to put our model to the test.

First, we need to update our model for the February data:

>>> model.partial_fit(X_feb, y_feb)

Next, we grab the March data and make our predictions, using the threshold of 1.12%:

>>> X_march, y_march = get_X_y(logs_2019, '2019-03', hackers_2019)>>> march_2019_preds = model.predict_proba(X_march)[:,1] >= threshold

Our classification ...