# amolgm/ufldl_selftaughtlearning architecture

> Architecture diagram of the GitHub repository amolgm/ufldl_selftaughtlearning, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-06-04). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/amolgm/ufldl_selftaughtlearning
- Repository: https://github.com/amolgm/ufldl_selftaughtlearning

## Overview

##### Project type and purpose
- This is a MATLAB-based machine-learning exercise (“SelfTaughtLearning”) implementing unsupervised feature learning and supervised classification on the MNIST dataset.

##### File structure
- Top level: README.md and a single folder `stl_exercise`.
- `stl_exercise/code`: all MATLAB `.m` files plus the `minFunc` subfolder. This is your core ML pipeline and optimizer library.
- `stl_exercise/code/minFunc`: a self-contained optimization toolkit (gradient, Hessian, line search routines) with both MATLAB and MEX/C implementations.
- `stl_exercise/data/mnist`: raw MNIST files (images and labels, both compressed and uncompressed).

##### README for additional insights
- The README confirms this is Andrew Ng’s “Self-taught learning” exercise.

## Diagram (Mermaid)

```mermaid
flowchart TB
    %% Orchestrator
    Orchestrator["stlExercise.m"]:::script
    click Orchestrator "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/stlExercise.m"

    %% Data Layer
    subgraph "Data Layer"
        direction TB
        MNISTFiles(("Raw MNIST Files")):::dataio
        click MNISTFiles "https://github.com/amolgm/ufldl_selftaughtlearning/tree/master/stl_exercise/data/mnist/"
        LoaderImages["loadMNISTImages.m"]:::dataio
        click LoaderImages "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/loadMNISTImages.m"
        LoaderLabels["loadMNISTLabels.m"]:::dataio
        click LoaderLabels "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/loadMNISTLabels.m"
    end

    %% Parameter Init
    InitParams["initializeParameters.m"]:::model
    click InitParams "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/initializeParameters.m"

    %% Unsupervised Feature Learner
    subgraph "Unsupervised Learning" 
        direction TB
        AutoencoderCost["sparseAutoencoderCost.m"]:::model
        click AutoencoderCost "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/sparseAutoencoderCost.m"
    end

    %% Optimizer
    subgraph "minFunc Optimizer" 
        direction TB
        MinFunc["minFunc.m"]:::optim
        click MinFunc "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/minFunc.m"
        Wolfe["WolfeLineSearch.m"]:::optim
        click Wolfe "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/WolfeLineSearch.m"
        Armijo["ArmijoBacktrack.m"]:::optim
        click Armijo "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/ArmijoBacktrack.m"
        LBFGS["lbfgs.m"]:::optim
        click LBFGS "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/lbfgs.m"
        AutoGrad["autoGrad.m"]:::optim
        click AutoGrad "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/autoGrad.m"
        AutoHess["autoHess.m"]:::optim
        click AutoHess "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/autoHess.m"
        AutoHv["autoHv.m"]:::optim
        click AutoHv "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/autoHv.m"
        AutoTensor["autoTensor.m"]:::optim
        click AutoTensor "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/autoTensor.m"
        LBFGSC["lbfgsC.c"]:::optim
        click LBFGSC "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/lbfgsC.c"
        MCHOLC["mcholC.c"]:::optim
        click MCHOLC "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/mcholC.c"
        Example1["example_minFunc.m"]:::optim
        click Example1 "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/example_minFunc.m"
        Example2["example_minFunc_LR.m"]:::optim
        click Example2 "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/minFunc/example_minFunc_LR.m"
        LogisticDir(("logistic/")):::optim
        click LogisticDir "https://github.com/amolgm/ufldl_selftaughtlearning/tree/master/stl_exercise/code/minFunc/logistic/"
    end

    %% Feature Extraction
    FeatureExtract["feedForwardAutoencoder.m"]:::model
    click FeatureExtract "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/feedForwardAutoencoder.m"

    %% Supervised Classifier
    subgraph "Supervised Classifier"
        direction TB
        SoftmaxTrain["softmaxTrain.m"]:::model
        click SoftmaxTrain "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/softmaxTrain.m"
        SoftmaxCost["softmaxCost.m"]:::model
        click SoftmaxCost "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/softmaxCost.m"
        SoftmaxPredict["softmaxPredict.m"]:::model
        click SoftmaxPredict "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/softmaxPredict.m"
    end

    %% Visualization
    Visualizer["display_network.m"]:::vis
    click Visualizer "https://github.com/amolgm/ufldl_selftaughtlearning/blob/master/stl_exercise/code/display_network.m"

    %% Connections
    Orchestrator --> MNISTFiles
    MNISTFiles --> LoaderImages
    MNISTFiles --> LoaderLabels
    LoaderImages --> InitParams
    LoaderLabels --> InitParams
    InitParams --> AutoencoderCost
    AutoencoderCost --> MinFunc
    MinFunc --> LBFGS
    MinFunc --> Wolfe
    MinFunc --> Armijo
    MinFunc --> AutoGrad
    MinFunc --> AutoHess
    MinFunc --> AutoHv
    MinFunc --> AutoTensor
    MinFunc --> LBFGSC
    MinFunc --> MCHOLC
    MinFunc --> Example1
    MinFunc --> Example2
    MinFunc --> LogisticDir
    MinFunc --> FeatureExtract
    FeatureExtract --> SoftmaxTrain
    SoftmaxTrain --> SoftmaxCost
    SoftmaxCost --> MinFunc
    SoftmaxTrain --> SoftmaxPredict
    SoftmaxPredict --> Orchestrator
    AutoencoderCost --> Visualizer
    FeatureExtract --> Visualizer

    %% Styles
    classDef dataio fill:#E0F7FA,stroke:#006064,color:#006064;
    classDef model fill:#E8F5E9,stroke:#1B5E20,color:#1B5E20;
    classDef optim fill:#FFF3E0,stroke:#E65100,color:#E65100;
    classDef vis fill:#F3E5F5,stroke:#4A148C,color:#4A148C;
    classDef script fill:#ECEFF1,stroke:#37474F,color:#37474F;
```
