Coursera machine learning week 6(二)
一 误差分析 当应用机器学习时，建议的顺序是： 1.首先构建一个比较简单的模型或学习算法； 2.画出模型的学习曲线、诊断模型是否有高偏差(欠拟合)或者高方差(过拟合)问题，从而决定是否需要更多的数据或者更多的特征来解决； 3.误差分析，手动检查算法算错的那些样例，然后找出算法算错的样例术语那种类型。 二 偏斜类的误差度量指标 2.1 偏斜类(Skewed classes)：是指假设一个样本中有两类
相关文章
- 1. Coursera Machine Learning Week 6 - Advice for Applying Machine Learning
- 2. Coursera machine learning week 6(一)
- 3. Sixth week of machine learning on Coursera
- 4. Third week of machine learning on Coursera
- 5. Seventh week of machine learning on Coursera
- 6. Second week of machine learning on Coursera
- 7. 【Machine Learning】【Andrew Ng】- Quiz(Week 6)
- 8. 机器学习系列之coursera week 6 Advice for Applying Machine Learning
- 9. 机器学习之Coursera Andrew Ng 《Machine Learning》 week 6 test 2
- 10. Machine Learning week 3
- 更多相关文章...
- • Docker Machine - Docker教程
- • Hibernate的二级缓存 - Hibernate教程
- • RxJava操作符（二）Transforming Observables
- • Kotlin学习（二）基本类型
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. Coursera Machine Learning Week 6 - Advice for Applying Machine Learning
- 2. Coursera machine learning week 6(一)
- 3. Sixth week of machine learning on Coursera
- 4. Third week of machine learning on Coursera
- 5. Seventh week of machine learning on Coursera
- 6. Second week of machine learning on Coursera
- 7. 【Machine Learning】【Andrew Ng】- Quiz(Week 6)
- 8. 机器学习系列之coursera week 6 Advice for Applying Machine Learning
- 9. 机器学习之Coursera Andrew Ng 《Machine Learning》 week 6 test 2
- 10. Machine Learning week 3