NanoPhasePredictor

NanoPhasePrediction is an AI-driven framework designed to predict material phases using a hybrid CNN-SVM (PPNN) approach. This project focuses on optimizing phase discovery in nano-scale phase prediction of PVDF through advanced machine learning techniques.

CNN-SVM Hybrid Model for Phase Prediction

This project implements a hybrid machine learning model combining a Convolutional Neural Network (CNN) for feature extraction and a Support Vector Machine (SVM) for classification. This hybrid approach leverages the CNN's ability to extract complex patterns from image data, while the SVM efficiently classifies the extracted features.

Overview

The CNN-SVM hybrid model is designed for tasks involving image classification. The CNN model processes input images, extracting high-level features. These features are then used as input for an SVM classifier to make the final predictions.

Requirements

To run the project, the following libraries are required:

• numpy
• tensorflow
• scikit-learn

Install the dependencies using:

pip install numpy tensorflow scikit-learn

Workflow

1. CNN for Feature Extraction:

   • The CNN architecture is defined in the create_cnn_model() function. It includes:
     • Two convolutional layers with ReLU activation and max-pooling layers.
     • A fully connected layer to reduce the feature dimensions.
     • A softmax output layer for initial classification.

2. Hybrid CNN-SVM Model:

   • After training the CNN using train_cnn_model(), the extracted features are passed to an SVM classifier. The SVM handles the final classification task using these extracted features.

3. Feature Extraction:

   • The CNN model is used to generate feature vectors from the input images using the extract_cnn_features() function.

4. SVM Training and Classification:

   • The SVM classifier is trained using the feature vectors extracted by the CNN. The classification performance is evaluated on the test data using accuracy as the metric.

Code Structure

• create_cnn_model(input_shape, num_classes): Defines the CNN architecture for feature extraction.
• train_cnn_model(X_train, y_train, input_shape, num_classes): Trains the CNN on the training dataset.
• extract_cnn_features(model, X): Extracts features from the CNN for classification.
• train_svm_model(X_train_features, y_train): Trains the SVM classifier using CNN-extracted features.
• main(): Main function to execute the entire hybrid pipeline from data generation, model training, feature extraction, SVM training, and evaluation.

How to Run

1. Download or clone this repository.
2. Ensure all dependencies are installed by running:

   pip install numpy tensorflow scikit-learn

3. Run the code:

   python Phase_prediction.ipynb

Model Training and Evaluation

• Feature Extraction: After training, the CNN is used to extract high-level features from the images, which serve as input to the SVM.
• SVM Training: The SVM classifier is trained on the features extracted from the CNN. The SVM uses a linear kernel.
• Model Evaluation: The accuracy of the hybrid CNN-SVM model is computed and displayed for the test dataset.

Author

https://github.com/ABnano