PeMSD4

The dataset refers to the traffic speed data in San Francisco Bay Area, containing 307 sensors on 29 roads. The time span of the dataset is January-February in 2018. It is a popular benchmark for traffic forecasting.

Benchmarks

Traffic Prediction/12 steps MAE Traffic Prediction/12 steps MAPE Traffic Prediction/12 steps RMSE

Related Benchmarks

PeMSD4 (10 days' training data, 15min)/Traffic Prediction/MAE PeMSD4 (10 days' training data, 15min)/Traffic Prediction/MAPE PeMSD4 (10 days' training data, 15min)/Traffic Prediction/RMSE PeMSD4 (10 days' training data, 30min)/Traffic Prediction/MAE PeMSD4 (10 days' training data, 30min)/Traffic Prediction/MAPE PeMSD4 (10 days' training data, 30min)/Traffic Prediction/RMSE PeMSD4 (10 days' training data, 60min)/Traffic Prediction/MAE PeMSD4 (10 days' training data, 60min)/Traffic Prediction/MAPE PeMSD4 (10 days' training data, 60min)/Traffic Prediction/RMSE