Emotion Classifier Application

Overview

This is a desktop application that classifies emotions from text input using a pre-trained Transformer model. The application utilizes tkinter for the GUI, Pillow for image handling, and transformers from Hugging Face for text classification.

Features

• Text Input: Users can input text into a text entry widget.
• Emotion Classification: Upon clicking the classify button, the application employs a Transformer model to classify the input text into different emotions.
• Result Display: The classified emotions are displayed along with their corresponding confidence scores, visually represented using progress bars.
• Clear and User-Friendly Interface: The GUI is simple and intuitive, featuring a header, text entry, and result display sections.

Requirements

• Python 3.6 or higher
• Required Python libraries: os, sys, tkinter, ttk, Pillow, transformers

Installation

1. Clone the Repository:

   git clone https://github.com/KCprsnlcc/TextClassificationAnalysis.git
   cd TextClassificationAnalysis

2. Install Dependencies:

   pip install -r requirements.txt

3. Ensure Model Directory: Ensure the pre-trained Transformer model is available in the transformermodel/ directory.

4. Place Logo: Ensure logo.png is in the base directory of the project.

Usage

1. Run the Application:

   python main.py

2. Enter Text: Type the text you want to classify into the text entry widget.

3. Classify Emotion: Click the "Classify Emotion" button to see the classified emotions along with confidence scores.

Code Explanation

Main Components

1. Import Statements: Import necessary libraries and modules.

   import os
   import sys
   import tkinter as tk
   from tkinter import ttk
   from PIL import Image, ImageTk
   from transformers import pipeline

2. Function Definitions:

   • classify_emotion: Classifies the input text and updates the results.
   • clear_results: Clears previous results.
   • display_input_text: Displays the input text.
   • update_results: Updates and displays new classification results.
   • get_progress_color: Returns the color for the progress bar based on the score.

3. Environment Configuration: Disable specific TensorFlow operations.

   os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'

4. Load Model: Load the emotion classification pipeline from the local directory.

   classifier = pipeline("text-classification", model=local_model_directory, top_k=None)

5. GUI Setup:

   • Main Window: Configure the main window.
   • Header Frame: Create a header frame with a logo and title.
   • Text Entry: Create a text entry widget for user input.
   • Classify Button: Create a button to trigger emotion classification.
   • Result Frame: Create a frame to display classification results.
   • Emotion Label: Label to display messages or status.

6. Start Main Event Loop: Start the tkinter main event loop to run the application.

   root.mainloop()

Acknowledgements

• Hugging Face for providing the pre-trained Transformer models.
• Python for being the programming language of choice.
• tkinter for the GUI framework.
• Pillow for image handling capabilities.