本文主要是介绍libcity 笔记：支持的模型，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！
1 支持的模型
1.1 traffic_state_pred
| HA | 历史平均值，将历史流量建模为季节性过程，然后使用前几个季节的加权平均值作为预测值。 |
| VAR | 向量自回归 |
| SVR | 支持向量回归 |
| ARIMA | |
| AutoEncoder | |
| Seq2Seq | 采用基于门控循环单元的编码器-解码器框架，进行多步预测 |
| FNN | 具有两个隐藏层和 L2 正则化的前馈神经网络 |
| RNN | |
1.1.1 交通流量预测 | |
| ACFM | 注意力人群流量机 Attentive Crowd Flow Machines, ACM Multimedia 2018 |
| MSTGCN | 降级版的ASTGCN，称为多组件时空图卷积网络，去掉了原模型的时空注意力机制 Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting AAAI 2019 |
| ASTGCN | Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting AAAI 2019 基于注意力的时空图卷积网络 |
| ST-ResNet | 时空残差网络 Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction AAAI 2017 |
| AGCRN | 自适应图卷积循环网络，通过自适应模块增强传统图卷积，并组合成循环神经网络，以捕捉细粒度时空关联。 Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting Neurips 2020 |
| Conv-GCN | 组合图卷积网络（GCN）与三维卷积神经网络（3D CNN） Multi-graph convolutional network for short-term passenger flow forecasting in urban rail transit. IET Intell. Trans. Syst.14, 10 (2020) |
| STDN | 时空动态网络（STDN）引入流门控机制以学习位置间的动态相似性 Revisiting Spatial-Temporal Similarity: A Deep Learning Framework for Traffic Prediction. AAAI 2019 |
| STSGCN | 时空同步图卷积网络（STSGCN） Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting. AAAI 2020 |
| ToGCN | 拓扑图卷积网络（ToGCN） Topological Graph Convolutional Network-Based Urban Traffic Flow and Density Prediction. IEEE Trans. Intell. Transp. Syst.(2020) |
| Multi-STGCnet | 含有三个作为时间组件的基于长短期记忆内存（LSTM）的模块和作为空间组件的三个用于提取目标站点空间关联的空间矩阵 Multi-STGCnet:A Graph Convolution Based Spatial-Temporal Framework for Subway PassengerFlow Forecasting IJCNN |
| ResLSTM | 合并残差网络（ResNet），图卷积网络（GCN）和长短期记忆内存（LSTM） Deep learning architecture for short-term passenger flow forecasting in urban rail transit. IEEE Trans. Intell. Transp. Syst.(2020) |
| CRANN | 可解释的、基于注意力的神经网络 A Spatio-Temporal Spot-Forecasting Framework forUrban Traffic Prediction ARXIV 2020 |
| DGCN | Dynamic Graph Convolution Network for Traffic Forecasting Based on Latent Network of Laplace Matrix Estimation. IEEE Trans. Intell. Transp. Syst.(2020) |
| DSAN | 动态切换注意力网络 Preserving Dynamic Attention for Long-Term Spatial-Temporal Prediction kdd 2020 |
| STNN | Spatio-Temporal Neural Networks for Space-Time Series Forecasting and Relations Discovery 2018 |
1.1.2 交通速度预测 | |
| DCRNN | Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting ICLR 2018 |
| STGCN | Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting IJCAI 2018 |
| GWNET | Graph Wave Net for Deep Spatial-Temporal Graph Modeling IJCAI 2019 |
| MTGNN | Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks. KDD 2020 |
| TGCN | T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction. IEEE Trans. Intell. Transp. Syst.21, 9 (2020) |
| TGCLSTM | Traffic Graph Convolutional Recurrent Neural Network: A Deep Learning Frame work for Network-Scale Traffic Learning and Forecasting. IEEE Trans. Intell. Transp.Syst.21, 11 (2020), |
| ATDM | On the Inclusion of Spatial Information for Spatio-Temporal Neural Networks 2020 |
| GMAN | GMAN:A Graph Multi-Attention Network for Traffic Prediction. AAAI 2020 |
| GTS | Discrete Graph Structure Learning forForecasting Multiple Time Series 2021 |
| STAGGCN | Spa-tiotemporal Adaptive Gated Graph Convolution Network for Urban Traffic FlowForecasting. CIKM 2020 |
| HGCN | 结构化图卷积网络 Hierarchical Graph Convolution Networks for Traffic Forecasting. (2021) |
| ST-MGAT | 时空多头图注意力机制网络，在图上直接建构卷积的同时，考虑邻居节点的特征和边权，生成新的节点表示 ST-MGAT:Spatial-Temporal Multi-Head Graph Attention Networks for Traffic Forecasting. In ICTAI 2020 |
| DKFN | 深度卡曼滤波网络 Graph Convolutional Networks with Kalman Filtering for Traffic Prediction SIGSPATIAL 2020 |
| STTN | Spatial-temporal transformer networks for traffic flow forecasting |
1.1.3 交通需求量预测 | |
| DMVSTNET | Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction AAAI 2018 |
| STG2Seq | STG2Seq: Spatial-Temporal Graph to Sequence Model for Multi-step Passenger Demand Forecasting. IJCAI 2019 |
| CCRNN | Coupled Layer-wise Graph Convolution for Transportation Demand Prediction AAAI 2021 |
1.1.4 OD预测 | |
| GEML | Origin-Destination Matrix Prediction via Graph Convolution: a New Perspective of Passenger Demand Modeling KDD 2019 |
| CSTN | Contextualized Spatial-Temporal Network for Taxi Origin-Destination Demand Prediction. In IEEE Transactions on Intelligent Transportation Systems. 2019 |
1.1.5 交通事故预测 | |
| GSNet | GSNet: Learning Spatial-Temporal Correlations from Geographical and Semantic Aspects for Traffic Accident Risk Forecasting AAAI 2021 |
1.2 traj_loc_pred
1.2.1 轨迹下一跳预测
| FPMC | 经典下一跳预测基线模型 Factorizing personalized Markov chains for next-basket recommendation. In WWW ACM 2010 |
| ST-RNN | 聚焦于在RNN隐藏层引入时空转移特性 Predicting the Next Location: A Recurrent Model with Spatial and Temporal Contexts AAAI 2016 |
| ATST-LSTM | 将轨迹点间的时间与距离差引入LSTM，并使用注意力机制 An attention-based spatiotemporal lstm network for next poi recommendation. IEEE Transactions on Services Computing(2019) |
| SERM | 在网络中引入轨迹的语义信息。SERM 模型依赖于 Glove 预训练语料库。 因此在使用该模型前，从Standard Dataset in LibCity - Google 云端硬盘下载了 |
| DeepMove | 混合历史和当前的轨迹进行预测，在这方面第一次采用注意力机制 DeepMove: Predicting Human Mobility with Attentional Recurrent Networks. In WWW. ACM 2018 |
| HST-LSTM | 将时空转移因子引入LSTM，并采用编码器-解码器架构进行预测 HST-LSTM: A Hierarchical Spatial-Temporal Long-Short Term Memory Network for Location Prediction. In IJCAI 2018 |
| LSTPM | 使用两个特别设计的LSTM捕捉用户长短期移动偏好，以联合二者预测下一位置 Where to Go Next: Modeling Long- and Short-Term User Preferences for Point-of-Interest Recommendation AAAI 2018 |
| GeoSAN | Geography-Aware Sequential Location Recommendation KDD 2020 |
| STAN | STAN: Spatio-Temporal Attention Network for Next Location Recommendation 2021 |
| CARA | A Contextual Attention Recurrent Architecture for Context-Aware Venue Recommendation. In SIGIR 2018 |
1.3 ETA
| DeepTTE | 端到端的深度学习模型，直接预测整条路经所需的旅行时间；提出了地理卷积操作，通过将地理信息整合到传统的卷积中，用来捕捉空间相关性 When Will You Arrive? Estimating Travel Time Based on Deep Neural Networks AAAI 2018 |
| TTPNet | 基于张量分解和图接入，可以从历史轨迹有效捕捉旅行速度和路网表征，以及可以更好地预测旅行时间 TTPNet: A Neural Network for Travel Time Prediction Based on Tensor Decomposition and Graph Embedding TKDE 2020 |
1.4 map_matching
| ST-Matching | Map-Matching for low-sampling-rate GPS trajectories. In: Proc. of the ACM-GIS. 2009 |
| IVMM | An interactive-voting based map matching algorithm. In: Proc. of the MDM. 2010 |
| HMMM | Hidden Markov map matching through noise and sparseness. In: Proc. of the ACM-GIS. 2009 |
1.5 road_representation
| ChebConv | 使用基于切比雪夫多项式近似的图卷积模型计算路网表征 Convolutional neural networks on graphs with fast localized spectral filtering NIPS 2016 |
| LINE | 适合大规模图结构的图嵌入模型，同时考虑一阶和二阶近似 Line: Large-scale information network embedding WWW 2015 |
| GeomGCN | 几何图神经网络 Geom-gcn: Geometric graph convolutional networks 2020 |
| DeepWalk | 将随机游走(random walk)和Word2Vec两种算法相结合的图结构数据挖掘算法 Deepwalk: Online learning of social representations KDD 2014 |
| Node2Vec | node2vec: Scalable feature learning for networks KDD 2016 |
| GAT | 图注意力网络 |
参考内容：复现模型列表 — Bigscity-LibCity 文档 (bigscity-libcity-docs.readthedocs.io)
这篇关于libcity 笔记：支持的模型的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！
