# iamdiwakaryan/ann_project architecture

> Architecture diagram of the GitHub repository iamdiwakaryan/ann_project, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-05-31). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/iamdiwakaryan/ann_project
- Repository: https://github.com/iamdiwakaryan/ann_project

## Overview

##### Project Type and Purpose
- This is a data-science / machine-learning pipeline project (“ann_project”) that trains and serves two neural-network models:
- A classification model (churn prediction)
- A regression model (salary prediction)
- The project includes raw data, preprocessing artifacts, training notebooks, TensorBoard logs, serialized model files, and an inference application (`app.py`).

##### File Structure Analysis
- Top-level artifacts:
- `Churn_Modelling.csv` – raw dataset for churn classification
- `main.ipynb`, `salaryregression.ipynb`, `prediction.ipynb` – Jupyter notebooks for EDA, model training, and prediction workflow
- Preprocessing pickles: `label_encoder.pkl`, `onehotencoder.pkl`, `scaler.pkl` (classification) and their `_reg` counterparts (regression)
- Trained models: `model.h5` (classification), `regerssion_model.h5` & `regerssion_model.keras` (regression)
- Logs: `logs/fit*` directories containing TensorBoard event files for training and validation runs
- `app.py` – inference script or micro-API
- `requirements.txt` – Python dependencies

##### Key Components
a. Data Ingestion
– Source: `Churn_Modelling.csv`
– Read by notebooks or `app.py` for batch training or live inference
b. Preprocessing Module
– Loads `LabelEncoder`, `OneHotEncoder`, `StandardScaler` pickles
– Transforms raw inputs into model-ready tensors
c. Training Pipelines
– Classification pipeline in `main.ipynb` → outputs `model.h5` + TensorBoard logs under `logs/fit*`
– Regression pipeline in `salaryregression.ipynb` → outputs `regerssion_model.h5`/`.keras` + logs
d. Model & Artifact Storage
– File-system bucket or local directory holding `.h5`, `.keras`, and `.pkl` files
e. Inference Service (`app.py`)
– Loads saved preprocessing artifacts and model files
– Exposes API or CLI endpoints for realtime predictions
f. Monitoring & Visualization
– TensorBoard reading from `logs/fit*` for training/validation metrics

##### Relationships & Data Flow
1) CSV data → Preprocessing → Model training
2) During training → Metrics emitted to TensorBoard logs
3) Trained model + preprocessors → persisted to disk
4) Inference service loads preprocessors & model → accepts new input → returns prediction

##### Architectural Patterns & Technologies
- Batch pipeline pattern for model training
- Artifact repository pattern for model/version management
- Frameworks: Python, TensorFlow/Keras for modeling; scikit-learn for preprocessing; TensorBoard for logging; Flask/FastAPI or pure script for `app.py`
- Storage: local filesystem for data, artifacts, and logs

## Diagram (Mermaid)

```mermaid
flowchart TD
    %% Data Ingestion
    subgraph "Data Ingestion"
        direction TB
        CSV["Churn_Modelling.csv"]:::data
    end

    %% Preprocessing Modules
    subgraph "Preprocessing Module" 
        direction TB
        subgraph "Classification"
            direction TB
            PC_label["label_encoder.pkl"]:::preprocessing
            PC_onehot["onehotencoder.pkl"]:::preprocessing
            PC_scaler["scaler.pkl"]:::preprocessing
        end
        subgraph "Regression"
            direction TB
            PR_label["label_encoder_reg.pkl"]:::preprocessing
            PR_onehot["onehotencoder_reg.pkl"]:::preprocessing
            PR_scaler["scaler_reg.pkl"]:::preprocessing
        end
    end

    %% Training Pipelines
    subgraph "Training Pipelines"
        direction TB
        subgraph "Classification Training"
            direction TB
            C_nb["main.ipynb"]:::training
            C_model["model.h5"]:::data
            C_log1["logs/fit20250207-162232"]:::monitor
            C_log2["logs/fit20250207-163106"]:::monitor
            C_log3["logs/fit20250207-221336"]:::monitor
        end
        subgraph "Regression Training"
            direction TB
            R_nb["salaryregression.ipynb"]:::training
            R_model1["regerssion_model.h5"]:::data
            R_model2["regerssion_model.keras"]:::data
            R_log["logs/fit_reg20250208-155137"]:::monitor
        end
    end

    %% Artifact & Dependency Management
    subgraph "Artifact & Dependency Management"
        direction TB
        Req["requirements.txt"]:::artifact
        Storage["Model & Preprocessor Storage"]:::data
    end

    %% Inference Service
    subgraph "Inference Service" 
        direction TB
        App["app.py"]:::service
        P_nb["prediction.ipynb"]:::service
        Output["Prediction Output"]:::service
    end

    %% Monitoring & Visualization
    subgraph "Monitoring & Visualization"
        direction TB
        TB["TensorBoard Dashboard"]:::monitor
    end

    %% Relationships
    CSV -->|reads data| PC_label
    CSV -->|reads data| PC_onehot
    CSV -->|reads data| PC_scaler
    CSV -->|reads data| PR_label
    CSV -->|reads data| PR_onehot
    CSV -->|reads data| PR_scaler

    PC_label -->|inputs| C_nb
    PC_onehot -->|inputs| C_nb
    PC_scaler -->|inputs| C_nb
    PR_label -->|inputs| R_nb
    PR_onehot -->|inputs| R_nb
    PR_scaler -->|inputs| R_nb

    C_nb -->|produces| C_model
    C_nb -->|logs metrics| C_log1
    C_nb -->|logs metrics| C_log2
    C_nb -->|logs metrics| C_log3

    R_nb -->|produces| R_model1
    R_nb -->|produces| R_model2
    R_nb -->|logs metrics| R_log

    C_model -->|stores| Storage
    R_model1 -->|stores| Storage
    R_model2 -->|stores| Storage
    PC_label -->|stores| Storage
    PC_onehot -->|stores| Storage
    PC_scaler -->|stores| Storage
    PR_label -->|stores| Storage
    PR_onehot -->|stores| Storage
    PR_scaler -->|stores| Storage

    Storage -->|loads artifacts| App
    PC_label -->|loads| App
    PC_onehot -->|loads| App
    PC_scaler -->|loads| App
    PR_label -->|loads| App
    PR_onehot -->|loads| App
    PR_scaler -->|loads| App

    App -->|calls| Output
    P_nb -->|orchestrates| App

    C_log1 -->|view| TB
    C_log2 -->|view| TB
    C_log3 -->|view| TB
    R_log -->|view| TB

    Req -->|dependencies| C_nb
    Req -->|dependencies| R_nb
    Req -->|dependencies| App

    %% Click Events
    click CSV "https://github.com/iamdiwakaryan/ann_project/blob/main/Churn_Modelling.csv"
    click PC_label "https://github.com/iamdiwakaryan/ann_project/blob/main/label_encoder.pkl"
    click PC_onehot "https://github.com/iamdiwakaryan/ann_project/blob/main/onehotencoder.pkl"
    click PC_scaler "https://github.com/iamdiwakaryan/ann_project/blob/main/scaler.pkl"
    click PR_label "https://github.com/iamdiwakaryan/ann_project/blob/main/label_encoder_reg.pkl"
    click PR_onehot "https://github.com/iamdiwakaryan/ann_project/blob/main/onehotencoder_reg.pkl"
    click PR_scaler "https://github.com/iamdiwakaryan/ann_project/blob/main/scaler_reg.pkl"
    click C_nb "https://github.com/iamdiwakaryan/ann_project/blob/main/main.ipynb"
    click C_log1 "https://github.com/iamdiwakaryan/ann_project/tree/main/logs/fit20250207-162232"
    click C_log2 "https://github.com/iamdiwakaryan/ann_project/tree/main/logs/fit20250207-163106"
    click C_log3 "https://github.com/iamdiwakaryan/ann_project/tree/main/logs/fit20250207-221336"
    click R_nb "https://github.com/iamdiwakaryan/ann_project/blob/main/salaryregression.ipynb"
    click R_model1 "https://github.com/iamdiwakaryan/ann_project/blob/main/regerssion_model.h5"
    click R_model2 "https://github.com/iamdiwakaryan/ann_project/blob/main/regerssion_model.keras"
    click R_log "https://github.com/iamdiwakaryan/ann_project/tree/main/logs/fit_reg20250208-155137"
    click App "https://github.com/iamdiwakaryan/ann_project/blob/main/app.py"
    click P_nb "https://github.com/iamdiwakaryan/ann_project/blob/main/prediction.ipynb"
    click Req "https://github.com/iamdiwakaryan/ann_project/blob/main/requirements.txt"
    click Storage "https://github.com/iamdiwakaryan/ann_project/blob/main/."
    
    %% Styles
    classDef data fill:#add8e6,stroke:#000;
    classDef preprocessing fill:#ccffcc,stroke:#000;
    classDef training fill:#ccffcc,stroke:#000;
    classDef service fill:#ffa500,stroke:#000;
    classDef monitor fill:#d3d3d3,stroke:#000;
    classDef artifact fill:#add8e6,stroke:#000;
```
