scikit-learn の使い方 (その4)

House Prices: Advanced Regression Techniques で 試してみました。

1) ライブラリーの読み込み
2) csv の　読み込み
3) model の作成　と　予測
4) csv の出力
の４ステップです。

1) ライブラリーの読み込み

[1]

import numpy as np
import pandas as pd

from sklearn.ensemble import RandomForestRegressor

2) csv の　読み込み

[2]

train_df=pd.read_csv('../input/house-prices-advanced-regression-techniques/train.csv')

test_df=pd.read_csv('../input/house-prices-advanced-regression-techniques/test.csv')
submission=pd.read_csv('../input/house-prices-advanced-regression-techniques/sample_submission.csv')

print(train_df.shape)
print(test_df.shape)
print(submission.shape)

3) model の作成　と　予測

[3]

predictor_cols = ['LotArea', 'OverallQual', 'YearBuilt', 'TotRmsAbvGrd']

train_x = train_df[predictor_cols]
train_y = train_df.SalePrice


model = RandomForestRegressor()
model.fit(train_x, train_y)

test_x = test_df[predictor_cols]

pred = model.predict(test_x)

print(pred[:20])

4) csv の出力

[4]

submission.SalePrice = pred
print(submission.head(20))
file_submit = "house_prices_submit.csv"
#
submission.to_csv(file_submit,index=False)