Kaggle实战：Digit Recognizer[Random Forest算法]

说明

正确率：94.014%，没有KNN效果好（96.800%），个人估计经过调参效果应该有所提升

代码

import pandas as pddata = pd.read_csv("train.csv")data.head()dataset = data.iloc[:,1:]   #提取特征dataset.head()label = data.iloc[:,0] #提取标签label.head()dataset.describe()label.describe()from sklearn.ensemble import RandomForestClassifierrf = RandomForestClassifier(oob_score=True,random_state=10)rf.fit(dataset, label)test = pd.read_csv("test.csv")pred = rf.predict(test)import numpy as npa = pd.Series(pred)b = pd.Series(np.arange(1,28000))c = pd.DataFrame([a,b])d = pd.DataFrame(c.T)d.to_csv("result.csv")