Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
WANGW09 4bce58537f | 2 years ago | |
---|---|---|
.META-INF | 2 years ago | |
PCL-SZ-volume | 2 years ago | |
data | 2 years ago | |
figures | 2 years ago | |
LRRCA.py | 2 years ago | |
LRRCA.tar.gz | 2 years ago | |
README.md | 2 years ago |
Linear Regression Regional Collaboration Algorithm for Traffic Flow Prediction
李佳栋
将深圳市交通流量数据集划分为训练集和测试集,实现线性回归式区域协同算法。
根据训练集优化算法参数,在测试集上进行预测,并计算MAE误差。
同时实现baseline算法(HA),通过统计和作图对比线性回归式区域协同算法相比HA方法在预测集上MAE的提升
Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20GHz
70s
MAE(Mean Absolute Error)
计算方法如下
Model | HA | LRRCA | Improvement |
---|---|---|---|
Global MAE | 18.74 | 14.65 | 21.82% |
源数据PCL-SZ-volume是通过处理深圳市交通委的数据得来的自有数据集,数据详情见./PCL-SZ-volume/PCL-SZ-volume.md
代码运行的环境与依赖。如下所示:
类别 | 名称 | 版本 |
---|---|---|
os | ubuntu | 16.04 |
numpy | 1.17.4 | |
pandas | 0.25.3 | |
matplotlib | 3.3.0 | |
scikit-learn |
代码的输入与输出。如下所示:
名称 | 说明 |
---|---|
输入 | h5格式交通流量数据集,采集点数为25632x136 |
输出 | 三个csv文件(真实值,baseline算法预测值,LRRCA算法预测值),小样本真实值和优化前后预测值的数值示例(若debug设为true),小样本真实值和优化前后预测值的曲线对比图(若fig设为true) |
在terminal下运行以下命令。
cd project_dir
python LRRCA.py --nid 20501805 --data PCL-SZ-volume/sz_cleaned.h5 --csv_path data --debug --fig
其中data
文件夹会存储所有的csv文件,grountruth是真实值,baseline是优化前,LRRCA是线性回归区域协同方法结果。figures
会存储可视化结果,名称与方法的对应同上。
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》