Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Source of ideas in this paper.

More »

Fig 1 Expand

Fig 2.

Visualization of traffic flow maps for TaxiBJ, BikeNYC datasets.

More »

Fig 2 Expand

Fig 3.

Urban gridding and definition of inflow and outflow.

More »

Fig 3 Expand

Fig 4.

Inflow flows in four zones over a week.

More »

Fig 4 Expand

Fig 5.

Comparison of traffic flow on sunny day and rainstorm.

More »

Fig 5 Expand

Fig 6.

The overall framework of the ST-D3DDARN model.

More »

Fig 6 Expand

Fig 7.

Comparison between 2D CNN and 3D CNN.

More »

Fig 7 Expand

Fig 8.

Comparison between ordinary 3D CNN and decoupled 3D CNN.

More »

Fig 8 Expand

Fig 9.

Decoupled 3D denseNet architecture.

More »

Fig 9 Expand

Fig 10.

ARN unit internal structure.

More »

Fig 10 Expand

Fig 11.

Spatial self-attention structure.

More »

Fig 11 Expand

Fig 12.

Coordinate attention mechanism structure.

More »

Fig 12 Expand

Fig 13.

External information branch structure.

More »

Fig 13 Expand

Fig 14.

Flow chart of traffic flow prediction in the whole city.

More »

Fig 14 Expand

Table 1.

Environment configuration.

More »

Table 1 Expand

Table 2.

Description of TaxiBJ and BikeNYC.

More »

Table 2 Expand

Fig 15.

Experimental results with different number of D3DDN layers.

(a) TaxiBJ. (b) BikeNYC.

More »

Fig 15 Expand

Fig 16.

Experimental results of different ARN module layers.

(a) TaxiBJ. (b) BikeNYC.

More »

Fig 16 Expand

Table 3.

Comparison of model characteristics.

More »

Table 3 Expand

Table 4.

Comparison of model prediction results.

More »

Table 4 Expand

Table 5.

Inflow and outflow prediction results in TaxiBJ.

More »

Table 5 Expand

Fig 17.

Error comparison of different models.

(a) TaxiBJ. (b) BikeNYC.

More »

Fig 17 Expand

Fig 18.

RMSE and MAE changes of the test set during the training of each model.

(a) TaxiBJ of RMSE. (b) TaxiBJ of MAE. (c) BikeNYC of MAE. (d) BikeNYC of MAE.

More »

Fig 18 Expand

Table 6.

Comparison of model efficiency.

More »

Table 6 Expand

Table 7.

Comparison of model variants of TaxiBJ.

More »

Table 7 Expand

Fig 19.

Performance of direct multi-step predictions for each method.

(a) Step-wise RMSE of TaxiBJ. (b) Step-wise MAE of TaxiBJ. (c) Step-wise RMSE of BikeNYC. (d) Step-wise MAE of BikeNYC.

More »

Fig 19 Expand

Fig 20.

Performance of recursive multi-step predictions for each method.

(a) Step-wise RMSE of TaxiBJ. (b) Step-wise MAE of TaxiBJ. (c) Step-wise RMSE of BikeNYC. (d) Step-wise MAE of BikeNYC.

More »

Fig 20 Expand