Practical Analysis of Evaluation Metrics in Classification Task

Classification is a supervised machine learning method that is often used in daily practice. There are various evaluation metrics that can be used to evaluate model performance in a particular classification task. This article will use a lithology classification case study to get to know some evaluation metrics that are often used, such as Accuracy Score, F1 Score, and MCC. First of all, we import the .CSV data which was subsetted from Poseidon Well Logs Dataset. To create a visualization, we need to convert the LITHO class to a number/code, so that it can be converted into a specific colour. From the code above, it can be seen that we have 3 litho classes that are imbalanced.