# yizhou-h/trajectory-mamba-cvpr architecture

> Architecture diagram of the GitHub repository yizhou-h/trajectory-mamba-cvpr, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-04-25). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/yizhou-h/trajectory-mamba-cvpr
- Repository: https://github.com/yizhou-h/trajectory-mamba-cvpr

## Overview

##### Main Components
a. Data Ingestion
- Argoverse1 and Argoverse2 datasets (images + annotations.json)
b. Preprocessing & Encoding
- Data loader modules
- Joint polyline encoding strategy
c. Core Model
- Encoder (selective SSM‐based attention layers)
- Decoder (trajectory generation head)
d. Configuration & Checkpoint Management
- configs/main_train.yml
- checkpoints/argoverseX_ckpt.pth
e. Evaluation & Output
- eval.py script
- GPU execution environment (CUDA_VISIBLE_DEVICES)

##### Relationships and Data Flow
- Data Ingestion → Preprocessing & Encoding
(polyline construction + feature embedding)
- Preprocessed Features → Encoder
(Selective State‐Space Model replaces quadratic self‐attention)
- Encoder Outputs → Decoder
(forecast future trajectories)
- Decoder Outputs → Evaluation Script
(compute metrics, export results)
- Config Files → Drive parameters for all pipeline stages
- Checkpoints → Model weight initialization before inference/evaluation

##### Architectural Patterns
- Encoder–Decoder Architecture
- Self‐Attention redesigned via Linear Selective SSM
- Joint Polyline Encoding for multi‐agent context
- Modular design (data, model, config, evaluation)

##### Call Out Critical Technologies & Frameworks
- PyTorch (model implementation)
- CUDA/NVIDIA GPUs (inference & training)
- Anaconda environment + environment.yaml
- YAML‐based configuration management
- Argoverse public datasets

– Clear component labels (e.g., “SelectiveSSMAttention”, “PolylineEncoder”)

## Diagram (Mermaid)

```mermaid
flowchart TD
    %% Data Layer
    subgraph "Data Layer"
        AV1["Argoverse1 Dataset"]:::data
        AV2["Argoverse2 Dataset"]:::data
    end

    %% Preprocessing & Encoding Layer
    subgraph "Preprocessing & Encoding"
        DataLoader["DataLoader Modules"]:::model
        Polyline["PolylineEncoder"]:::model
    end

    %% Model Layer
    subgraph "Model Layer"
        Enc["Encoder"]:::model
        SSM["SelectiveSSMAttention"]:::model
        Dec["Decoder"]:::model
    end

    %% Config & Checkpoint Layer
    subgraph "Config & Checkpoint"
        ConfigFile["configs/main_train.yml"]:::config
        Checkpoint["checkpoints/argoverseX_ckpt.pth"]:::config
        Gitignore[".gitignore"]:::config
        README["README.md"]:::config
    end

    %% Evaluation & Output Layer
    subgraph "Evaluation & Output"
        EvalScript["eval.py"]:::eval
        GPU["GPU Execution Environment"]:::eval
    end

    %% Data Flow
    AV1 -->|"raw images & annotations"| DataLoader
    AV2 -->|"raw images & annotations"| DataLoader
    DataLoader -->|"polyline construction & feature embedding"| Polyline
    Polyline -->|"embedded features"| Enc
    Enc -->|"context representations"| SSM
    SSM -->|"selective SSM outputs"| Dec
    Dec -->|"predicted trajectories"| EvalScript
    EvalScript -->|"runs on"| GPU

    %% Config & Checkpoint Dependencies
    ConfigFile -->|drives parameters| DataLoader
    ConfigFile -->|drives parameters| Enc
    ConfigFile -->|drives parameters| Dec
    Checkpoint -->|weight initialization| Dec

    %% Click Events
    click README "https://github.com/yizhou-h/trajectory-mamba-cvpr/blob/main/README.md"
    click Gitignore "https://github.com/yizhou-h/trajectory-mamba-cvpr/blob/main/.gitignore"

    %% Styles
    classDef data fill:#ADD8E6,stroke:#333,stroke-width:1px
    classDef model fill:#90EE90,stroke:#333,stroke-width:1px
    classDef config fill:#FFA500,stroke:#333,stroke-width:1px
    classDef eval fill:#DDA0DD,stroke:#333,stroke-width:1px
```
