how can i choose the accuracy scorer ?

Choosing the appropriate accuracy scorer depends on the specific problem you are trying to solve and the nature of the data. Here are some general guidelines to help you select an appropriate accuracy scorer:

1. Classification vs Regression: If you are working on a classification problem, you might use accuracy metrics such as precision, recall, F1 score, or AUC-ROC. On the other hand, if you are working on a regression problem, you might use metrics such as mean squared error (MSE), root means squared error (RMSE), or mean absolute error (MAE).
2. Balanced vs Imbalanced Datasets: If your dataset is balanced (i.e., a roughly equal number of samples in each class), you can use accuracy as your primary evaluation metric. However, if your dataset is imbalanced (i.e., there are significantly more samples in one class than the other), you may need to use additional evaluation metrics such as precision, recall, or F1 score to avoid being misled by a high accuracy score that is primarily due to the majority class.
3. Problem-specific considerations: Some problems may require specific evaluation metrics. For example, if you are working on a recommendation system, you might use metrics such as mean average precision (MAP) or normalized discounted cumulative gain (NDCG).
4. Your specific objectives: Your choice of accuracy metric should align with your specific objectives. For example, if your goal is to minimize false positives, you might prioritize precision over recall and vice versa.

In summary, selecting an appropriate accuracy scorer depends on the specific problem you are working on, the nature of the data, and your specific objectives. It's always a good idea to review the relevant literature and consult with domain experts to choose the most suitable accuracy scorer for your task.