美文网首页
【ML】Metrics

【ML】Metrics

作者: 盐果儿 | 来源:发表于2023-04-12 19:23 被阅读0次

    Metric of Machine Learning

    Accuracy

    \text{Accuracy} = \frac{\text{Number of correctly classified instances}}{\text{Total number of instances}}

    from sklearn.metrics import accuracy_score

    accuracy = accuracy_score(y_test, y_pred)


    A confusion matrix is not a metric in machine learning, but it is a useful tool for understanding the performance of a classification model and calculating evaluation metrics.

    Precision

    Recall

    F1-score

    ROC-AUC score

    Mean Squared Error

    Root Mean Squared Frror

    R-squared score

    Log loss

    Metrics of Knowledge Graph Embedding:

    HITS@1

    HITS@3

    HITS@10

    HIT(Hit Rate,HR)命中率:它反映的是在推荐序列中是否包含了用户真正点击的item。

    MR

    MRR(Mean Reciprocal Rank,MRR)平均倒数排名,该指标反应的是我们找到的这些item是否摆在用户更明显的位置,强调位置关系,顺序性。

    相关文章

      网友评论

          本文标题:【ML】Metrics

          本文链接:https://www.haomeiwen.com/subject/mzmirdtx.html