(可视化模型表现)Visualizing Model Performance
来源:互联网 发布:如何设计一个软件 编辑:程序博客网 时间:2024/05/16 19:54
- Stakeholders outside of the data science team may have little patience for details, and will often want a higher-level, more intuitive view of model performance.
- It is important for the data scientist to realize that clear communication with key stakeholders is not only a primary goal of her job, but also is essential for doing the right modeling (in addition to doing the modeling right).
————————————————————————————————————————————————
主要内容:
- Profit curves
- ROC graph (receiver operating characteristic)
- AUC graph (The area under the ROC Curve)
- Cumulative response and lift curve
————————————————————————————————————————————————
(一)Profit curve
The disadvantage of a profit graph is that it requires that operating conditions be known and specified exactly. With many real-world problems, the operating conditions are imprecise or change over time, and the data scientist must contend with uncertainty.
(二)ROC graph
在先验分布和损益不清楚的情况下,我们可以依据不同情况依次建立模型绘制Profit graph,但这显然不能让人满意
为解决这个问题,我们可以将各种情况的模型表现都绘制在一张图中,其中一种方法便是 Receiver Operating Characteristics (ROC) graph
ROC graph 的详细产生过程见下图
ROC数据点的计算方式
Confusion Matrix
P
N
Ya
b
N
c
d
ROC Point
An advantage of ROC graphs is that they decouple classifier performance from the conditions under which the classifiers will be used. Specifically, they are independent of the class proportions as well as the costs and benefits.
The area under the RCO curve (AUC)
- Though a ROC curve provides more information than its area, the AUC is useful when a single number is needed to summarize performance, or when nothing is known about the operating conditions.
(三)Cumulative response and lift curve
- (可视化模型表现)Visualizing Model Performance
- maven之可视化项目依赖(Visualizing dependencies)
- 使用View Model从表现层分离领域模型
- 使用View Model从表现层分离领域模型
- 模型(Model)类
- Django06模型(Model)
- Caffe 的可视化 (三) caffe model 的可视化
- CSS学习5(视觉表现模型)
- Process performance Model
- Visualizing Representations: Deep Learning and Human Beings 简单翻译(数据可视化:深度学习和人类)(未完)
- 域模型(Domain Model)
- 盒模型(BOX Model)
- 瀑布模型(Waterfall Model)
- 增量模型(Incremental Model)
- 螺旋模型(Spiral Model)
- 演化模型(Evolutionary Model)
- SIPOC模型(SIPOC Model)
- 模型验证(Model Validation)
- spring 解读之 基本bean装配
- 在Mathematica中定义矩阵的半张量积运算
- 使用adb解锁
- CF #244 (Div. 2)D
- bzoj 3238: [Ahoi2013]差异
- (可视化模型表现)Visualizing Model Performance
- jQury animate操作 background-position 方法
- Android之TextView实现跑马灯的效果(笔记)
- C#求平均分和高于平均分的同学姓名。
- linux下动态库的符号冲突、隐藏和强制优先使用库内符号
- [Hnoi2013]游走(bzoj3143)
- java通过HTTP接收xml和返回xml
- POJ 3111 K Best(最大化平均值)
- zoj2112(单点修改区间第K小)