# rezahabibi96/tscil-ext architecture

> Architecture diagram of the GitHub repository rezahabibi96/tscil-ext, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-05-21). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/rezahabibi96/tscil-ext
- Repository: https://github.com/rezahabibi96/tscil-ext

## Overview

##### Project type and purpose
- This is an open-source machine-learning pipeline (not a web app) focused on time-series class-incremental learning.
- It implements a family of continual-learning “agents” (including the novel GCPP prototype-based generative classifier) on various datasets, orchestrates experiments (via scripts and Ray Tune), trains models with PyTorch, buffers samples for replay, and produces visualizations.

##### File structure
- Top-level directories:
- agents/ → implementations of each continual-learning strategy (ER, DER, LwF, EWC, GCPP, etc.)
- data/ → dataset wrappers and preprocessing for UCI-HAR, UWAVE, WISDM, etc.
- models/ → neural architectures (encoder, classifier, TimeVAE generator, normalization layers)
- utils/ → shared utilities: replay buffers, data streaming, metrics, optimizers, name matching
- experiment/ → orchestration code: exp.py (runs experiments), tune_and_exp.py (Ray Tune entry), plus configuration files
- shell/ → bash wrappers for batch runs per dataset
- Entry points: main_config.py (standard experiments), main_tune.py (hyperparameter search with Ray).

##### README
- Confirms continual-learning on time series, lists supported datasets, algorithms (GCPP + many baselines), dependencies (Python 3.10, PyTorch 1.13.1, Ray 2.3.1).
- Acknowledges external frameworks influencing design: TSCIL, Avalanche, Mammoth, etc., indicating plugin-style integration.

##### Capturing essential details
- Break out Buffer sub-modules: retrieval vs update strategies.
- Base classes (agents/base.py, models/base.py).
- How configuration files drive dynamic instantiation.

Use a top-down layout:
- Top row: Configuration & Orchestrator → reads Config, invokes Data & Agents
- Middle row (three lanes):
– Data Lane: data/ → preprocessing → DataLoader
– Training Lane: Agent + Model + Buffer → Training Loop → Metrics
– Visualization Lane: plot/ → Reports
- Side annotations: PyTorch compute engine, Ray cluster, filesystem (datasets, checkpoints).

## Diagram (Mermaid)

```mermaid
flowchart TB
    %% Orchestration & Configuration
    subgraph "Orchestration & Configuration"
        direction TB
        Config["tune_config.py\n(Configuration)"]:::orchestration
        ExpPy["exp.py\n(Experiment Orchestrator)"]:::orchestration
        TuneExp["tune_and_exp.py\n(Ray Tune Entry)"]:::orchestration
        MainCfg["main_config.py\n(Standard Entry)"]:::orchestration
        MainTune["main_tune.py\n(Hyperparam Entry)"]:::orchestration
    end

    %% Data Lane
    subgraph "Data Lane"
        direction TB
        DataDir["data/\n(Dataset Wrappers)"]:::data
        DataUtil["utils/data.py\n(Data Utils)"]:::data
        DataLoader["Data Loader"]:::data
    end

    %% Training Lane
    subgraph "Training Lane"
        direction TB
        AgentBase["agents/base.py\n(Agent Base)"]:::training
        AgentsDir["agents/\n(Specific Agents)"]:::training
        ModelBase["models/base.py\n(Model Base)"]:::training
        ModelsDir["models/\n(Architectures)"]:::training
        NormDir["models/normalization/\n(Norm Layers)"]:::training
        Classifier["models/classifier.py"]:::training
        Encoders["models/encoders.py"]:::training
        TimeVAE["models/timeVAE.py"]:::training
        BufferDir["utils/buffer/\n(Buffer Storage)"]:::training
        BufferFile["utils/buffer/buffer.py"]:::training
        BufferUtils["utils/buffer/buffer_utils.py"]:::training
        TrainingLoop["Training Loop"]:::training
        Metrics["Metrics & Logs"]:::training
    end

    %% Visualization Lane
    subgraph "Visualization Lane"
        direction TB
        PlotDir["plot/\n(Visualization Scripts)"]:::viz
        Plotter["Plotter"]:::viz
    end

    %% External Services
    subgraph "External Services"
        direction TB
        RayCluster["Ray Cluster"]:::external
        PyTorch["PyTorch Engine"]:::external
        FileSystem["Filesystem\n(Datasets & Checkpoints)"]:::external
    end

    %% Connections
    Config -->|reads| ExpPy
    Config -->|reads| TuneExp
    Config -->|reads| MainCfg
    Config -->|reads| MainTune

    ExpPy -->|invokes| DataDir
    TuneExp -->|invokes| DataDir
    MainCfg -->|invokes| DataDir
    MainTune -->|invokes| DataDir

    DataDir -->|uses| DataUtil
    DataUtil -->|produces| DataLoader

    DataLoader -->|feeds| AgentsDir
    AgentsDir -->|extends| AgentBase
    AgentsDir -->|instantiates| ModelsDir
    ModelsDir -->|extends| ModelBase
    ModelsDir -->|includes| Classifier
    ModelsDir -->|includes| Encoders
    ModelsDir -->|includes| TimeVAE
    NormDir -->|used by| ModelsDir

    AgentsDir -->|uses| BufferDir
    BufferDir -->|contains| BufferFile
    BufferDir -->|contains| BufferUtils

    AgentsDir -->|runs| TrainingLoop
    TrainingLoop -->|calls| PyTorch
    TrainingLoop -->|stores/retrieves| BufferDir
    TrainingLoop -->|emits| Metrics

    Metrics -->|read by| Plotter
    Plotter -->|loads| PlotDir

    ExpPy -->|dispatches| RayCluster
    TuneExp -->|dispatches| RayCluster
    MainTune -->|dispatches| RayCluster

    DataDir -->|loads/saves| FileSystem
    BufferDir -->|persists| FileSystem
    ModelsDir -->|saves checkpoints| FileSystem
    Metrics -->|logs| FileSystem

    %% Click Events
    click DataDir "https://github.com/rezahabibi96/tscil-ext/tree/main/data/"
    click DataUtil "https://github.com/rezahabibi96/tscil-ext/blob/main/utils/data.py"
    click AgentsDir "https://github.com/rezahabibi96/tscil-ext/tree/main/agents/"
    click AgentBase "https://github.com/rezahabibi96/tscil-ext/blob/main/agents/base.py"
    click ModelsDir "https://github.com/rezahabibi96/tscil-ext/tree/main/models/"
    click ModelBase "https://github.com/rezahabibi96/tscil-ext/blob/main/models/base.py"
    click Classifier "https://github.com/rezahabibi96/tscil-ext/blob/main/models/classifier.py"
    click Encoders "https://github.com/rezahabibi96/tscil-ext/blob/main/models/encoders.py"
    click TimeVAE "https://github.com/rezahabibi96/tscil-ext/blob/main/models/timeVAE.py"
    click NormDir "https://github.com/rezahabibi96/tscil-ext/tree/main/models/normalization/"
    click BufferDir "https://github.com/rezahabibi96/tscil-ext/tree/main/utils/buffer/"
    click BufferFile "https://github.com/rezahabibi96/tscil-ext/blob/main/utils/buffer/buffer.py"
    click BufferUtils "https://github.com/rezahabibi96/tscil-ext/blob/main/utils/buffer/buffer_utils.py"
    click ExpPy "https://github.com/rezahabibi96/tscil-ext/blob/main/experiment/exp.py"
    click TuneExp "https://github.com/rezahabibi96/tscil-ext/blob/main/experiment/tune_and_exp.py"
    click MainCfg "https://github.com/rezahabibi96/tscil-ext/blob/main/main_config.py"
    click MainTune "https://github.com/rezahabibi96/tscil-ext/blob/main/main_tune.py"
    click Config "https://github.com/rezahabibi96/tscil-ext/blob/main/experiment/tune_config.py"
    click PlotDir "https://github.com/rezahabibi96/tscil-ext/tree/main/plot/"

    %% Styles
    classDef data fill:#ADD8E6,stroke:#000,stroke-width:1px
    classDef training fill:#90EE90,stroke:#000,stroke-width:1px
    classDef orchestration fill:#FFA500,stroke:#000,stroke-width:1px
    classDef viz fill:#DDA0DD,stroke:#000,stroke-width:1px
    classDef external fill:#EEEEEE,stroke:#000,stroke-width:1px
```
