Week 4 of #mlzoomcamp was all about ML Evaluation
The lessons covered Evaluation Metrics on classification models.
After training a model, it’s performance needs to be evaluated on a test set. This helps to understand how well the model will generalize on a new data.
There are a number of different evaluation metrics that we can use for binary classification problems.
Some of the most common evaluation metrics and concepts include:
☑ Accuracy
☑ Confusion Matrix
☑ Precision
☑ Recall
☑ Class Imbalance and it's importance
☑ F1 Score
☑ Receiver Operating Characteristic Area Under the Curve (ROC AUC).
☑ ROC Curve
☑ K-Fold Cross Validation
The goal of the homework was to apply the evaluation metrics on the classification problem (Bank Marketing dataset - desired target for classification task will be the 'converted' variable - has the client signed up to the platform or not?) from Week 3
Top comments (0)