Traffic Prediction Benchmarks¶
We provide benchmark results of spatiotemporal prediction learning (STL) methods on popular traffic prediction datasets. More STL methods will be supported in the future. Issues and PRs are welcome! Currently, we only provide benchmark results, trained models and logs will be released soon (you can contact us if you require these files). You can download model files from Baidu Cloud (3t2t).
Table of Contents¶
Currently supported spatiotemporal prediction methods
[x] ConvLSTM (NeurIPS’2015)
[x] PredNet (ICLR’2017)
[x] PredRNN (NeurIPS’2017)
[x] PredRNN++ (ICML’2018)
[x] E3D-LSTM (ICLR’2018)
[x] MIM (CVPR’2019)
[x] CrevNet (ICLR’2020)
[x] PhyDNet (CVPR’2020)
[x] MAU (NeurIPS’2021)
[x] PredRNN.V2 (TPAMI’2022)
[x] SimVP (CVPR’2022)
[x] SimVP.V2 (ArXiv’2022)
[x] TAU (CVPR’2023)
[x] DMVFN (CVPR’2023)
Currently supported MetaFormer models for SimVP
[x] ViT (ICLR’2021)
[x] Swin-Transformer (ICCV’2021)
[x] MLP-Mixer (NeurIPS’2021)
[x] ConvMixer (Openreview’2021)
[x] UniFormer (ICLR’2022)
[x] PoolFormer (CVPR’2022)
[x] ConvNeXt (CVPR’2022)
[x] VAN (ArXiv’2022)
[x] IncepU (SimVP.V1) (CVPR’2022)
[x] gSTA (SimVP.V2) (ArXiv’2022)
[x] HorNet (NeurIPS’2022)
[x] MogaNet (ArXiv’2022)
TaxiBJ Benchmarks¶
We provide traffic benchmark results on the popular TaxiBJ dataset using \(4\rightarrow 4\) frames prediction setting. Metrics (MSE, MAE, SSIM, pSNR) of the best models are reported in three trials. Parameters (M), FLOPs (G), and V100 inference FPS (s) are also reported for all methods. All methods are trained by Adam optimizer with Cosine Annealing scheduler (5 epochs warmup and min lr is 1e-6) and single GPU.
STL Benchmarks on TaxiBJ¶
For a fair comparison of different methods, we report the best results when models are trained to convergence. We provide config files in configs/taxibj.
Method |
Setting |
Params |
FLOPs |
FPS |
MSE |
MAE |
SSIM |
PSNR |
Download |
---|---|---|---|---|---|---|---|---|---|
ConvLSTM-S |
50 epoch |
14.98M |
20.74G |
815 |
0.3358 |
15.32 |
0.9836 |
39.45 |
|
E3D-LSTM* |
50 epoch |
50.99M |
98.19G |
60 |
0.3427 |
14.98 |
0.9842 |
39.64 |
|
PhyDNet |
50 epoch |
3.09M |
5.60G |
982 |
0.3622 |
15.53 |
0.9828 |
39.46 |
|
PredNet |
50 epoch |
12.5M |
0.85G |
5031 |
0.3516 |
15.91 |
0.9828 |
39.29 |
|
PredRNN |
50 epoch |
23.66M |
42.40G |
416 |
0.3194 |
15.31 |
0.9838 |
39.51 |
|
MIM |
50 epoch |
37.86M |
64.10G |
275 |
0.3110 |
14.96 |
0.9847 |
39.65 |
|
MAU |
50 epoch |
4.41M |
6.02G |
540 |
0.3268 |
15.26 |
0.9834 |
39.52 |
|
PredRNN++ |
50 epoch |
38.40M |
62.95G |
301 |
0.3348 |
15.37 |
0.9834 |
39.47 |
|
PredRNN.V2 |
50 epoch |
23.67M |
42.63G |
378 |
0.3834 |
15.55 |
0.9826 |
39.49 |
|
DMVFN |
50 epoch |
3.54M |
0.057G |
6347 |
3.3954 |
45.52 |
0.8321 |
31.14 |
|
SimVP+IncepU |
50 epoch |
13.79M |
3.61G |
533 |
0.3282 |
15.45 |
0.9835 |
39.45 |
|
SimVP+gSTA-S |
50 epoch |
9.96M |
2.62G |
1217 |
0.3246 |
15.03 |
0.9844 |
39.71 |
|
TAU |
50 epoch |
9.55M |
2.49G |
1268 |
0.3108 |
14.93 |
0.9848 |
39.74 |
Benchmark of MetaFormers on SimVP (MetaVP)¶
Similar to Moving MNIST Benchmarks, we benchmark popular Metaformer architectures on SimVP with training times of 50-epoch. We provide config files in configs/taxibj/simvp.
MetaFormer |
Setting |
Params |
FLOPs |
FPS |
MSE |
MAE |
SSIM |
PSNR |
Download |
---|---|---|---|---|---|---|---|---|---|
SimVP+IncepU |
50 epoch |
13.79M |
3.61G |
533 |
0.3282 |
15.45 |
0.9835 |
39.45 |
|
SimVP+gSTA-S |
50 epoch |
9.96M |
2.62G |
1217 |
0.3246 |
15.03 |
0.9844 |
39.71 |
|
ViT |
50 epoch |
9.66M |
2.80G |
1301 |
0.3171 |
15.15 |
0.9841 |
39.64 |
|
Swin Transformer |
50 epoch |
9.66M |
2.56G |
1506 |
0.3128 |
15.07 |
0.9847 |
39.65 |
|
Uniformer |
50 epoch |
9.52M |
2.71G |
1333 |
0.3268 |
15.16 |
0.9844 |
39.64 |
|
MLP-Mixer |
50 epoch |
8.24M |
2.18G |
1974 |
0.3206 |
15.37 |
0.9841 |
39.49 |
|
ConvMixer |
50 epoch |
0.84M |
0.23G |
4793 |
0.3634 |
15.63 |
0.9831 |
39.41 |
|
Poolformer |
50 epoch |
7.75M |
2.06G |
1827 |
0.3273 |
15.39 |
0.9840 |
39.46 |
|
ConvNeXt |
50 epoch |
7.84M |
2.08G |
1918 |
0.3106 |
14.90 |
0.9845 |
39.76 |
|
VAN |
50 epoch |
9.48M |
2.49G |
1273 |
0.3125 |
14.96 |
0.9848 |
39.72 |
|
HorNet |
50 epoch |
9.68M |
2.54G |
1350 |
0.3186 |
15.01 |
0.9843 |
39.66 |
|
MogaNet |
50 epoch |
9.96M |
2.61G |
1005 |
0.3114 |
15.06 |
0.9847 |
39.70 |
|
TAU |
50 epoch |
9.55M |
2.49G |
1268 |
0.3108 |
14.93 |
0.9848 |
39.74 |