Linux Productivity Apps

(For Students, Programmers, and Data's People)

1. Flow Chart Maker

For people who want chatgpt to generate flow charts.

How to Generate

1. Prompt GPT or any other LLM to writemake a .dot file
2. Run the app executable and either browse for .dot file, or paste.dot file content here

Sample Prompt

I’m designing a flowchart to explain a process. Please generate a valid Graphviz DOT file (not an image, only the DOT text) using clear structure and labels.

Title: “Student Research Workflow”
Style: simple, left-to-right layout (rankdir=LR), rounded boxes, arrows showing data flow.

Steps to include:
- Define Research Question
- Collect Data
- Clean Data
- Analyze Results
- Write Report

3. Select the location where you want to save .PNG Flowchart File.

---

Example

Input

digraph T5 {
    rankdir=LR;
    fontsize=12;
    labelloc="t";
    label="T5 Architecture (Encoder–Decoder with High & Low Level Abstractions)";

    // Node styles
    node [shape=box, style=filled, fontname="Helvetica"];

    // Input Embeddings
    input [label="Input Tokens\n(Embeddings + Positional Info)", fillcolor="#cce5ff"];

    // Encoder
    subgraph cluster_encoder {
        label="Encoder Stack (N Layers)";
        style=filled;
        color="#99ccff";
        
        enc_in [label="Input Embedding Vector", fillcolor="#e6f2ff"];
        
        subgraph cluster_enc_layer {
            label="Encoder Layer";
            color="#b3d9ff";
            style=filled;

            enc_self_attn [label="Multi-Head Self-Attention", fillcolor="#ffcccb"];
            enc_norm1 [label="Layer Norm + Residual", fillcolor="#ffe6e6"];
            enc_ffn [label="Feed-Forward Network", fillcolor="#ccffcc"];
            enc_norm2 [label="Layer Norm + Residual", fillcolor="#e6ffe6"];
        }
        
        enc_out [label="Encoder Hidden States\n(Meaning Soup)", fillcolor="#e6f2ff"];
        
        enc_in -> enc_self_attn -> enc_norm1 -> enc_ffn -> enc_norm2 -> enc_out;
    }

    // Decoder
    subgraph cluster_decoder {
        label="Decoder Stack (N Layers)";
        style=filled;
        color="#ffcc99";
        
        dec_in [label="Shifted Output Embeddings", fillcolor="#fff0e6"];
        
        subgraph cluster_dec_layer {
            label="Decoder Layer";
            color="#ffd9b3";
            style=filled;
            
            dec_self_attn [label="Masked Multi-Head Self-Attention", fillcolor="#ffcccb"];
            dec_norm1 [label="Layer Norm + Residual", fillcolor="#ffe6e6"];
            dec_cross_attn [label="Cross-Attention\n(Q=Decoder, K,V=Encoder)", fillcolor="#ffeb99"];
            dec_norm2 [label="Layer Norm + Residual", fillcolor="#fff7cc"];
            dec_ffn [label="Feed-Forward Network", fillcolor="#ccffcc"];
            dec_norm3 [label="Layer Norm + Residual", fillcolor="#e6ffe6"];
        }
        
        dec_out [label="Decoder Hidden States", fillcolor="#fff0e6"];
        
        dec_in -> dec_self_attn -> dec_norm1 -> dec_cross_attn -> dec_norm2 -> dec_ffn -> dec_norm3 -> dec_out;
    }

    // Output
    output [label="Softmax\n(Output Tokens)", fillcolor="#d5f5e3"];

    // Connections
    input -> enc_in;
    enc_out -> dec_cross_attn;
    dec_out -> output;
}

Output

---

2. Local OCR

For people who needs offline image to text converter.

Usage Guide

• You can load image from file menu or just directly paste it from clipboard.
• Then click perform OCR button and text from image will appear.

Note : For better quality use high quality image

---

How to create executable files and run on your systems.

Requirements

• Python 3.8+
• PyQt6
• Pillow
• pytesseract
• Graphviz (for Flowchart Maker)
• Tesseract OCR (for OCR Tool)

Install all Python deps in projects .venv:

pip install PyQt6 pillow pytesseract

System Libraries

Linux (Ubuntu / Fedora)

Use apt or dnf to install system packages

# Ubuntu
sudo apt install graphviz tesseract-ocr

# Fedora
sudo dnf install graphviz tesseract

Build Executables :

In your projects virtual environment install pyinstaller and create executable like that :

pip install pyinstaller
pyinstaller --onefile --windowed flow_chart_maker/flow_chart_maker.py
pyinstaller --onefile --windowed OCR/main.py

For Windows

Windows

Install Python. Install Graphviz. Install Tessaract OCR.

Add all to path.

then follow the same process as linux.

MacOS

Use brew to install system packages :

brew install python3 graphviz tesseract
pip3 install PyQt6 pillow pytesseract

Follow the same process as in linux for creating executable files.