[英]Error metric of h2o random forest explanation
I am running h2o random forest with the following parameter setting 我正在使用以下参数设置运行h2o随机森林
model_rf <- h2o.randomForest(x = predictors, y = labels,
training_frame = train_data, classification = T,
importance = T,
verbose = T, type = "BigData", ntree = 50)
After running I am getting the following output. 运行后,我得到以下输出。
Model Details:
==============
H2ORegressionModel: drf
Model ID: DRFModel__906d074da6ebf8057525b2b61c1c4c87
Model Summary:
number_of_trees model_size_in_bytes min_depth max_depth mean_depth min_leaves max_leaves mean_leaves
1 50.000000 2708173.000000 20.000000 20.000000 20.00000 4200.000000 5241.000000 4720.70000
H2ORegressionMetrics: drf
** Reported on training data. **
Description: Metrics reported on Out-Of-Bag training samples
MSE: 0.0006302392
R2 : -0.03751038
Following are my questions. 以下是我的问题。
1) What does MSE and R2 mean? 1)MSE和R2是什么意思?
2) If they are mean square error or similar why am I getting these metric for a classification setting? 2)如果它们是均方误差或相似的值,为什么要为分类设置获取这些度量?
3) How do I get other metrics like gini or auc? 3)如何获得其他指标,如基尼系数或auc?
4) Can i say that if these 2 params decrease with a different parameter setting, my model performance has improved? 4)我可以说如果这两个参数随着参数设置的不同而减少,我的模型性能会得到改善吗?
Here are the answers to your questions: 1. MSE stands for mean squared error. 以下是对您的问题的答案:1. MSE代表均方误差。 Essentially it measures the difference between the estimator and the estimated.R2 measures how well-fit a statistical model is. 本质上,它衡量的是估计量与估计值之间的差异.R2衡量的是统计模型的拟合度。
Using MSE you can judge how often you model misclassified data. 使用MSE,您可以判断对错误分类的数据进行建模的频率。
If you are using Flow, click on Inspect and then output-training_metrics to see MSE, R2, AUC, gini, etc. 如果您正在使用Flow,请单击“ 检查” ,然后单击output-training_metrics以查看MSE,R2,AUC,gini等。
Sorry, I'm not sure I understand this question. 抱歉,我不确定我是否理解这个问题。 Are you asking if a decreaed gini or AUC equate to improved model performance? 您是否在问降低基尼系数或AUC是否等同于改善模型性能?
Avni 阿夫尼
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.