【Kaggle】【Titanic】【AutoGluon】测试

安装

!pip install autogluon
#安装autogluon

训练

from autogluon.tabular import TabularDataset,TabularPredictor
#训练
train_data = TabularDataset('../input/titanic/train.csv')
id,label = 'PassengerId','Survived'
predictor = TabularPredictor(label=label).fit(train_data.drop(columns=[id]))

#查看训练数据
train_data.head(30)
print(train_data.shape)

TabularDataset读取训练数据
提取’PassengerId’,'Survived’分别作为"id",“label”
TabularPredictor对“Survived”进行预测，fit除了“id”列的train_data

预测

import pandas as pd
#预测
test_data = TabularDataset('../input/titanic/test.csv')
preds = predictor.predict(test_data.drop(columns=[id]))
submission = pd.DataFrame({id:test_data[id],label:preds})
submission.to_csv('submission.csv',index=False)

#查看测试数据
test_data.head(30)
print(test_data.shape)

TabularDataset读取测试数据
predictor.predict预测除了“id”列的test_data
pandas合并test_data的"id"列与预测值
to_csv导出结果

提交方式

Competition首页

Notebook首页

结果

这里排名动态变化

Reference

10行代码战胜90%数据科学家？–跟李沐学AI–Bilibili