Python-sklearn机器学习的第一个样例(5)
来源:互联网 发布:软件方案评审结论 编辑:程序博客网 时间:2024/06/11 03:59
In [61]:
iris_data_clean = pd.read_csv('iris-data-clean.csv')# We're using all four measurements as inputs# 注意到 scikit-learn 要求所有的记录都要用“列表”list形式表示, e.g.,# [ [val1, val2, val3],# [val1, val2, val3],# ... ]# 所以,我们要把输入数据集转化为一个列表的列表(a list of lists)# We can extract the data in this format from pandas like this:all_inputs = iris_data_clean[['sepal_length_cm', 'sepal_width_cm', 'petal_length_cm', 'petal_width_cm']].values# Similarly, we can extract the classesall_classes = iris_data_clean['class'].values# Make sure that you don't mix up the order of the entries# all_inputs[5] inputs should correspond to the class in all_classes[5]# Here's what a subset of our inputs looks like:all_inputs[:5]
Out[61]:
In [62]:
from sklearn.cross_validation import train_test_split(training_inputs, testing_inputs, training_classes, testing_classes) = train_test_split(all_inputs, all_classes, train_size=0.75, random_state=1)
In [63]:
from sklearn.tree import DecisionTreeClassifier# Create the classifierdecision_tree_classifier = DecisionTreeClassifier()# Train the classifier on the training setdecision_tree_classifier.fit(training_inputs, training_classes)# Validate the classifier on the testing set using classification accuracydecision_tree_classifier.score(testing_inputs, testing_classes)
Out[63]:
In [64]:
model_accuracies = []for repetition in range(1000): (training_inputs, testing_inputs, training_classes, testing_classes) = train_test_split(all_inputs, all_classes, train_size=0.75) decision_tree_classifier = DecisionTreeClassifier() decision_tree_classifier.fit(training_inputs, training_classes) classifier_accuracy = decision_tree_classifier.score(testing_inputs, testing_classes) model_accuracies.append(classifier_accuracy) sb.distplot(model_accuracies)
Out[64]:
阅读全文
0 0
- Python-sklearn机器学习的第一个样例(5)
- Python-sklearn 机器学习的第一个样例(1)
- Python-sklearn机器学习的第一个样例(2)
- Python-sklearn机器学习的第一个样例(3)
- Python-sklearn机器学习的第一个样例(4)
- Python-sklearn机器学习的第一个样例(6)
- Python-sklearn 机器学习的第一个样例(7)
- Python机器学习库SKLearn包含的内容(目录)
- python sklearn 机器学习库的安装
- 基于python的机器学习库Sklearn
- [机器学习]基于python的机器学习库Sklearn-01
- python 机器学习-sklearn基本功能
- python机器学习(1)python.sklearn库
- 安装Python的机器学习包Sklearn 出错解决方法
- python中sklearn机器学习实现的博客
- sklearn:Python语言开发的通用机器学习库
- Python机器学习包的sklearn中的Gridsearch简单使用
- Python机器学习库SKLearn的特征选择
- 如何在linux桌面上创建webstorm快捷方式
- 用指针和数组实现字符串复制
- 九度oj 题目1007:奥运排序问题
- 计算有限5位数
- BZOJ 4896 Thu Summer Camp2016 补退选 Trie树
- Python-sklearn机器学习的第一个样例(5)
- SpringMVC学习记录(四)--文件上传与下载
- 【等你来战】京东金融-中信证券杯量化策略大赛第6期!
- poj1321之DFS
- |BZOJ 4034|树链剖分|线段树|[HAOI2015]树上操作
- MPI Matrix Multiplication
- 关于Android studio 中无法打开ddms的data目录
- 第六次上级作业
- 【算法作业13】LeetCode 338. Counting Bits