QFNN

URL: https://github.com/prayagtiwari/QFNN

Description:
This repository implements a Quantum Fuzzy Neural Network (QFNN) for multimodal sentiment and sarcasm detection. The project combines quantum computing with fuzzy neural networks to handle the complexity of multimodal data, specifically targeting sentiment and sarcasm analysis.

Methods

The repository includes the following main components:

• ExpDatasets.py: Dataloader for the dataset.
• hyper_mt.py: Implements the multimodal fusion part of QFNN.
• qModule.py: Implements the quantum composition of QFNN, combining quantum computing techniques with fuzzy logic for enhanced model performance.

Results

Specific performance metrics are not detailed in the repository description. The methodology focuses on combining quantum and fuzzy techniques for sentiment and sarcasm detection.

Dataset

No specific dataset is mentioned in the repository, but it is designed for multimodal sentiment and sarcasm detection, which suggests the use of datasets that include text, images, or other modalities.

How to Run

1. Run hyper_mt.py to pre-train the multimodal fusion part of QFNN.
2. Run qModule.py to train the entire QFNN.
3. Run expressibility.ipynb to test the expressibility of parametrized quantum circuits.
4. Run entanglement.ipynb to test the entanglement capability of quantum circuits.