🧠 SimpleRNN

A simple Recurrent Neural Network (RNN) built using Keras to perform sentiment analysis on the IMDB movie reviews dataset.

---

📦 Setup Locally

Get started in minutes:

1. Clone the repository

git clone https://github.com/coder-tejas/SimpleRNN.git
cd SimpleRNN

2. (Optional) Set up a virtual environment

# Create virtual environment
python -m venv venv

# Activate
# On Windows
venv\Scripts\activate

# On Linux/Mac
source venv/bin/activate

3. Install dependencies

pip install -r requirements.txt

4. Run the project

Launch Jupyter Notebooks:

jupyter notebook

Open and run:

• RNN.ipynb
• embedding.ipynb
• simplernn.ipynb
• prediction.ipynb

Or run the standalone Python script:

python main.py

(⚡ Make sure simple_rnn_imdb.h5 model file is present if you want to load a pretrained model.)

---

🛠️ Project Structure

SimpleRNN/
├── RNN.ipynb            # Introduction to RNNs
├── embedding.ipynb      # Word embedding demo
├── simplernn.ipynb      # Building/training SimpleRNN
├── prediction.ipynb     # Predictions using the model
├── main.py              # Script version
├── simple_rnn_imdb.h5   # Pretrained model file
├── requirements.txt     # List of dependencies
└── README.md            # (This file!)

---

🧰 Tech Stack

• Python 3.7+
• TensorFlow & Keras
• NumPy
• Jupyter Notebook

---

🚀 Future Improvements

• Add LSTM / GRU models for better performance
• Explore attention mechanisms
• Fine-tune with custom datasets

---

📝 License

This project is licensed under the MIT License.

---

✨ Acknowledgments

Built with ❤️ by coder-tejas