🧠 CNN Text Classifier Assistant – Streamlit App

This is an interactive Streamlit chatbot application that guides users through building a CNN-based text classification model using a dataset they upload. The app leverages OpenAI's GPT-4o for intelligent guidance and performs data preprocessing, model training, and confusion matrix evaluation—all in a user-friendly, conversational format.

---

📁 Project Structure

├── .gitignore            # Ignore saved models, temporary files, etc.
├── app.py                # Main Streamlit application
├── requirements.txt      # Python dependencies

---

🚀 Features

• ✅ Upload CSV dataset
• 🧹 Text preprocessing with customizable options
• 🔠 Tokenization & label encoding for CNN input
• 🧠 Dynamic CNN model training with adjustable architecture
• 📊 Confusion Matrix visualization after training
• 💾 Model download (as .zip) post training
• 🤖 OpenAI GPT-4o integration for interactive guidance

---

🔧 Setup Instructions

1. Clone the repository

git clone https://github.com/yourusername/cnn-text-classifier-chatbot.git
cd cnn-text-classifier-chatbot

2. Create and activate a virtual environment (optional but recommended)

python -m venv venv
source venv/bin/activate    # On Windows use `venv\Scripts\activate`

3. Install dependencies

pip install -r requirements.txt

4. Set your OpenAI API key

You can either:

• Export it in your terminal:

export OPENAI_API_KEY=your_openai_api_key

• Or modify app.py to hardcode it temporarily:

api_key = "your_openai_api_key"

5. Run the app

streamlit run app.py

---

📌 How to Use

1. Upload a CSV file via the sidebar.

2. Select the text and label columns.

3. Follow the chatbot's guidance to:

   • Explore data
   • Clean and preprocess it
   • Prepare it for CNN training
   • Configure and train the model

4. Visualize the confusion matrix.

5. Download the trained model.

---

🧠 Model Architecture

The CNN model is built dynamically based on user-defined layers (Conv1D, MaxPooling1D, Dense, Dropout, etc.). Users can customize:

• Number of layers
• Kernel sizes, filters
• Activation functions
• Optimizer, learning rate, and more

---

📦 Dependencies

Major libraries:

• streamlit
• openai
• pandas, numpy, matplotlib
• tensorflow, sklearn

Full list in requirements.txt

---

📈 Example Use Case

Upload a dataset of product reviews labeled as "positive" or "negative", clean the data, define a CNN model, train it, and visualize classification performance—all via a conversational assistant.

---

---

🪪 License

MIT License (or your own if different)