Neural Evolution Engine V3.0 is an advanced Deep Learning application built from scratch using TensorFlow and Keras. It simulates evolutionary biology principles by predicting the optimal adaptation strategy for a species when faced with catastrophic environmental threats.
What's New in V3.0? (The Cognitive Leap!) V3.0 features a revolutionary Triple-Branch Multi-Modal Architecture that handles three distinct tasks: processing Biological Data, analyzing Environmental Imagery, and providing Natural Language assistance via a chatbot.
The system utilizes a sophisticated architecture that combines two main model structures for prediction and one for conversational AI.
| Branch | Architecture | Input | Function |
|---|---|---|---|
| Visual Branch (The "Eye") | CNN (2x Conv2D + MaxPooling) | 64x64 RGB Images | Analyzes visual patterns (snow, fire, toxic waste) to identify the threat. |
| Biological Branch (The "Brain") | ANN (Dense Layers) | Encoded Biological Features | Processes organism's physiological constraints and traits. |
| Fusion Layer | Concatenate + Softmax | Merged CNN/ANN features | Generates probability distribution for 6 evolutionary outcomes. |
| Branch | Architecture | Input | Function |
|---|---|---|---|
| Language Branch | LSTM (Recurrent Neural Network) | User Text (Tokenized, Embedded) | Understands questions about Evolutionary Biology (e.g., "What is genetic drift?"). |
| Function | Intent Classification | Chatbot Model (evolutionchatbotmodel.h5) |
Provides relevant, pre-trained biological explanations. |
When running main.py, the user is presented with THREE MAIN OPTIONS:
| Option | Mode | Primary AI Used | Description |
|---|---|---|---|
| 1 | Simulation Mode (Hybrid) | CNN + ANN | Predicts the optimal evolutionary adaptation based on visual threats and organism traits. |
| 2 | AI Assistant Mode | LSTM (NLP) | Answers user queries regarding evolutionary concepts and biological definitions. |
| 3 | Quit | - | Exits the program. |
- Deep Learning: TensorFlow, Keras (Functional API)
- Sequence Modeling: LSTM (New!)
- Computer Vision: OpenCV (Image Preprocessing)
- Data Engineering: Pandas, NumPy
- Preprocessing: Scikit-Learn (StandardScaler, LabelEncoder)
- Clone the Repository
git clone [https://github.com/ralolooafanxyaiml/neural-evolution-engine]
cd Neural-Evolution-Engine
pip install tensorflow pandas numpy scikit-learn opencv-python- Train the NLP Chatbot Model (One Time Setup)
python chatbot_train.py- Run the Main Engine
python main.pyData Sources & Acknowledgements This project utilizes external datasets for training the Visual Threat Detection (CNN) module:
Intel Image Classification by Puneet Bansal (Cold/Ice)
Natural Disaster Images by Aseem Arora (Heat/Fire)
Garbage Classification by Sashaank Sekar (Toxin/Pollution)
Underwater Image Classification by Great Sharma (Airless/Aquatic)
US Drought Data (Scarcity)
Developed by Mustafa İlker Aktaş - Global AI Contributor