【Kaggle笔记】Bike Sharing Demand

最新推荐文章于 2023-07-28 21:08:22 发布

原创最新推荐文章于 2023-07-28 21:08:22 发布 · 1.3k 阅读

2 ·

CC 4.0 BY-SA版权

文章标签：

#数据

Python 专栏收录该内容

81 篇文章

订阅专栏

本文介绍了一个自行车租赁需求预测的比赛项目，使用随机森林回归算法对自行车租赁需求进行了预测，并详细展示了从数据读取到特征工程再到模型训练及结果输出的全过程。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

比赛题目

https://www.kaggle.com/c/bike-sharing-demand

代码

# -*- coding:utf-8 -*-

import pandas as pd

# 读入原始数据
train_df = pd.read_csv("train.csv", header=0)
test_df = pd.read_csv("test.csv", header=0)


# 选取特征值
selected_features = ['datetime', 'season', 'holiday', 'workingday', 'weather', 'temp', 'atemp', 'humidity', 'windspeed']

X_train_df = train_df[selected_features]
y_train_df = train_df['count']

X_test_df = test_df[selected_features]


# 特征值处理
X_train_df['month'] = pd.DatetimeIndex(X_train_df.datetime).month
X_train_df['day'] = pd.DatetimeIndex(X_train_df.datetime).dayofweek
X_train_df['hour'] = pd.DatetimeIndex(X_train_df.datetime).hour
X_train_df = X_train_df.drop(['datetime'], axis=1)  # axis=1意思为每行执行，axis=0表示每列执行

X_test_df['month'] = pd.DatetimeIndex(X_test_df.datetime).month
X_test_df['day'] = pd.DatetimeIndex(X_test_df.datetime).dayofweek
X_test_df['hour'] = pd.DatetimeIndex(X_test_df.datetime).hour
X_test_df = X_test_df.drop(['datetime'], axis=1)


# 采用DictVectorizer进行特征向量化
from sklearn.feature_extraction import DictVectorizer
dict_vec = DictVectorizer(sparse=False)

X_train_df = dict_vec.fit_transform(X_train_df.to_dict(orient='record'))
X_test_df = dict_vec.transform(X_test_df.to_dict(orient='record'))


# 使用RandomForestRegressor进行回归预测
# from sklearn.ensemble import GradientBoostingRegressor
from sklearn.ensemble import RandomForestRegressor
gbr = RandomForestRegressor()
gbr.fit(X_train_df, y_train_df)
gbr_y_predict = gbr.predict(X_test_df)


# 输出结果
gbr_submission = pd.DataFrame({'datetime': test_df['datetime'], 'count': gbr_y_predict})
gbr_submission.to_csv('gbr_submission.csv', index=False)