# omid-ghozatlou/query-by-example-using-svdd architecture

> Architecture diagram of the GitHub repository omid-ghozatlou/query-by-example-using-svdd, 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/omid-ghozatlou/query-by-example-using-svdd
- Repository: https://github.com/omid-ghozatlou/query-by-example-using-svdd

## Overview

##### Project type and purpose
- This is a research‐oriented deep learning toolkit for “query by example” image retrieval in remote sensing archives, built around an enhanced Deep SVDD (Support Vector Data Description) framework.
- Key capabilities: multiple neural architectures (Gabor CA ResNet, Split Transform Network, various LeNets), a TCAL (Triple Criteria Active Learning) extension, dataset loaders/preprocessors, training pipelines, and evaluation/visualization tools (t-SNE, histograms).

##### File structure
- Top-level scripts / entry points:
- main.py, script.py – generic runners
- deepSVDD.py – core SVDD class
- TCAL.py – active‐learning extension
- Gabor_CA_ResNet.py, Split_Transform_Network.py – network definitions
- IR_evaluation.py, tsne.py, histogram.py – evaluation/analysis
- base/ – abstract base classes for datasets, networks, trainers
- datasets/ – each file handles a dataset (CIFAR-10, EuroSAT, MNIST, etc.) + preprocessing
- networks/ – concrete network architectures ready to plug into DeepSVDD or TCAL
- optim/ – trainer implementations (ae_trainer, deepSVDD_trainer)
- utils/ – config loading, result collection, visualization helpers
- notebooks/ – example Jupyter notebooks demonstrating usage

##### Extract insights from the README
- The workflow:
a. Load images via dataset modules
b. Preprocess & augment
c. Instantiate a network (e.g., SplitTransformNetwork)
d. Wrap it in DeepSVDD (or TCAL for active learning)
e. Train via optim/*.py trainers
f. Retrieve/query images and evaluate with IR_evaluation, t-SNE, histograms
- Tech stack: Python 3.7+, PyTorch, NumPy, scikit-learn, Matplotlib

##### Components & relationships
a. Main components (blocks):
1. Data Sources – CIFAR-10, EuroSAT, MNIST, custom datasets
2. Preprocessing Pipeline – normalization, augmentations, transforms
3. Network Architectures –
- Gabor Convolutional Attention ResNet
- Split Transform Network
- LeNet variants (cifar10_LeNet, scatter_LeNet*)
4. Model Wrappers – DeepSVDD, TCAL
5. Trainers – AETrainer, DeepSVDDTrainer
6. Retrieval/Evaluation – IR_evaluation, tsne, histogram
7. Visualization – plot_images_grid, t-SNE scatter plots, histograms
8. Configuration & Utilities – config parsing, results logging
9. Entry Points – main.py, script.py, notebooks

b. Data flow & interactions:
- Data Sources → Preprocessing → Network Architectures
- Networks + hyperparameters → wrapped by Model Wrappers
- Model Wrappers + Trainers → Training Loop → Checkpoints / Metrics
- Trained Model + Query Image → Retrieval Module → Ranked Results
- Results → Evaluation Scripts (IR_evaluation, tsne, histogram)
- Evaluation Outputs → Visualization Module → Plots

c. Architectural patterns / principles:
– Modular design via base classes (open/closed principle): plug‐in new datasets, nets, trainers
– Separation of concerns: data loading, model definition, training logic, evaluation, visualization
– Configurable pipelines via utils/config.py
– Reproducibility: clear script/notebook entry points

d. Key technologies & frameworks:
– PyTorch (nn.Modules in networks/, training in optim/)
– NumPy / scikit-learn (some preprocessing, metrics, t-SNE)
– Matplotlib (visualization)
– Python scripts + Jupyter notebooks for interactive demo

##### Final advice
- Then expand each stage into subcomponents, reflecting the directory structure.
- Review with the team to ensure that **extension points** and **dependency flows** are correctly captured.

## Diagram (Mermaid)

```mermaid
flowchart TB
    %% Data Sources
    subgraph "Data Sources" 
        direction TB
        CIFAR10["CIFAR-10 Dataset"]:::data
        EuroSAT["EuroSAT Dataset"]:::data
        EuroSAT_MS["EuroSAT Multispectral"]:::data
        MNIST["MNIST Dataset"]:::data
        FashionMNIST["Fashion MNIST"]:::data
        OtherDatasets["Other Datasets"]:::data
    end

    %% Preprocessing
    Preprocessing["Preprocessing Pipeline\n(normalization, augmentations)"]:::data

    %% Network Architectures
    subgraph "Network Architectures"
        direction TB
        GaborCA["Gabor CA ResNet"]:::model
        SplitTransform["Split Transform Network"]:::model
        ResNetBackbone["ResNet Backbone"]:::model
        LeNetVariants["LeNet Variants"]:::model
    end

    %% Model Wrappers
    subgraph "Model Wrappers"
        direction TB
        DeepSVDD["DeepSVDD Wrapper"]:::model
        TCAL["TCAL Active Learning"]:::model
    end

    %% Trainers
    subgraph "Training Components"
        direction TB
        AETrainer["Autoencoder Trainer"]:::train
        SVDDTrainer["DeepSVDD Trainer"]:::train
    end

    %% Retrieval & Evaluation
    subgraph "Retrieval & Evaluation"
        direction TB
        IREval["IR_evaluation"]:::eval
        TSNE["t-SNE Analysis"]:::eval
        Histogram["Histogram Analysis"]:::eval
    end

    %% Visualization
    Visualization["Visualization Helpers\n(image grids, plots)"]:::vis

    %% Configuration & Utilities
    subgraph "Configuration & Utilities"
        direction TB
        Config["Config Parser"]:::util
        CollectResults["Results Logging"]:::util
    end

    %% Entry Points
    subgraph "Entry Points"
        direction TB
        MainPy["main.py"]:::util
        ScriptPy["script.py"]:::util
        Notebook["notebooks/example.ipynb"]:::util
    end

    %% Pipeline Flow
    CIFAR10 --> Preprocessing
    EuroSAT --> Preprocessing
    EuroSAT_MS --> Preprocessing
    MNIST --> Preprocessing
    FashionMNIST --> Preprocessing
    OtherDatasets --> Preprocessing

    Preprocessing --> GaborCA
    Preprocessing --> SplitTransform
    Preprocessing --> ResNetBackbone
    Preprocessing --> LeNetVariants

    GaborCA --> DeepSVDD
    SplitTransform --> DeepSVDD
    ResNetBackbone --> DeepSVDD
    LeNetVariants --> DeepSVDD

    DeepSVDD --> AETrainer
    DeepSVDD --> SVDDTrainer
    TCAL --> SVDDTrainer

    AETrainer --> IREval
    SVDDTrainer --> IREval

    IREval --> Visualization
    TSNE --> Visualization
    Histogram --> Visualization

    Config -.-> Preprocessing
    Config -.-> GaborCA
    Config -.-> DeepSVDD
    Config -.-> AETrainer
    CollectResults -.-> SVDDTrainer

    MainPy -.-> Config
    ScriptPy -.-> Config
    Notebook -.-> Config
    Notebook -.-> Visualization

    %% Click Events
    click CIFAR10 "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/datasets/cifar10.py"
    click EuroSAT "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/datasets/eurosat.py"
    click EuroSAT_MS "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/datasets/eurosat_MS.py"
    click MNIST "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/datasets/mnist.py"
    click FashionMNIST "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/datasets/mnist_Fashion.py"
    click Preprocessing "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/datasets/preprocessing.py"
    click GaborCA "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/Gabor_CA_ResNet.py"
    click SplitTransform "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/Split_Transform_Network.py"
    click ResNetBackbone "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/networks/ResNet.py"
    click LeNetVariants "https://github.com/omid-ghozatlou/query-by-example-using-svdd/tree/master/networks/"
    click DeepSVDD "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/deepSVDD.py"
    click TCAL "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/TCAL.py"
    click AETrainer "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/optim/ae_trainer.py"
    click SVDDTrainer "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/optim/deepSVDD_trainer.py"
    click IREval "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/IR_evaluation.py"
    click TSNE "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/tsne.py"
    click Histogram "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/histogram.py"
    click Visualization "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/utils/visualization/plot_images_grid.py"
    click Config "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/utils/config.py"
    click CollectResults "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/utils/collect_results.py"
    click MainPy "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/main.py"
    click ScriptPy "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/script.py"
    click Notebook "https://github.com/omid-ghozatlou/query-by-example-using-svdd/blob/master/notebooks/example.ipynb"

    %% Styles
    classDef data fill:#D0E8FF,stroke:#004080
    classDef model fill:#DFF2BF,stroke:#4F8A10
    classDef train fill:#FFE599,stroke:#BF5F00
    classDef eval fill:#E2D5F5,stroke:#5D3A9B
    classDef vis fill:#FDE9D9,stroke:#CC4125
    classDef util fill:#F0F0F0,stroke:#888888
```
