# Using Row Weights to Penalize False Negatives

Example of using row weights to penalize false negatives

In many cases, penalty for misclassifying a sample is not symmetric between positives and negatives. For example, when diagnosing a disease, a false negative can be much worse than false positive. This example shows how to create a vector of row weights from class labels: positive sample is given 10x weight compared to negative sample. The model is then trained using the weighted data.