| # RoBERTa Fine-Tuned Model for Question Answering | |
| This repository hosts a fine-tuned version of the RoBERTa model optimized for question-answering tasks using the [SQuAD](w) dataset. The model is designed to efficiently perform question answering while maintaining high accuracy. | |
| ## Model Details | |
| - **Model Architecture**: RoBERTa | |
| - **Task**: Question Answering | |
| - **Dataset**: [SQuAD](w) (Stanford Question Answering Dataset) | |
| - **Quantization**: FP16 | |
| - **Fine-tuning Framework**: Hugging Face Transformers | |
| ## 🚀 Usage | |
| ### Installation | |
| ```bash | |
| pip install transformers torch | |
| ``` | |
| ### Loading the Model | |
| ```python | |
| from transformers import RobertaTokenizer, RobertaForQuestionAnswering | |
| import torch | |
| device = "cuda" if torch.cuda.is_available() else "cpu" | |
| model_name = "AventIQ-AI/roberta-chatbot" | |
| model = RobertaForQuestionAnswering.from_pretrained(model_name).to(device) | |
| tokenizer = RobertaTokenizer.from_pretrained(model_name) | |
| ``` | |
| ### Chatbot Inference | |
| ```python | |
| from transformers import pipeline | |
| # Load QA pipeline | |
| qa_pipeline = pipeline("question-answering", model=model, tokenizer=tokenizer, device=0) | |
| # Test sample question | |
| # Updated context and question for a flight prediction example | |
| # Updated context and question for flight prediction example | |
| context = "Flight AI101 departs from New York at 10:00 AM and arrives in San Francisco at 1:30 PM. The flight duration is 5 hours and 30 minutes." | |
| question = "What is the duration of Flight AI101?" | |
| # Get answer | |
| result = qa_pipeline(question=question, context=context) | |
| print(result) | |
| ``` | |
| 📊 Evaluation Results | |
| After fine-tuning the RoBERTa-base model for question answering, we evaluated the model's performance on the validation set from the SQuAD dataset. The following results were obtained: | |
| Metric Score Meaning | |
| F1 Score 89.5 Measures the balance between precision and recall for answer extraction. | |
| Exact Match 82.4 Percentage of questions where the predicted answer matches the ground truth exactly. | |
| Fine-Tuning Details | |
| Dataset | |
| The SQuAD dataset, containing over 100,000 question-answer pairs based on Wikipedia articles, was used for fine-tuning the model. | |
| Training | |
| Number of epochs: 3 | |
| Batch size: 8 | |
| Evaluation strategy: steps | |
| Quantization | |
| Post-training quantization was applied using PyTorch's built-in quantization framework to reduce the model size and improve inference efficiency. | |
| Repository Structure | |
| ``` | |
| . | |
| ├── model/ # Contains the quantized model files | |
| ├── tokenizer_config/ # Tokenizer configuration and vocabulary files | |
| ├── model.safetensors/ # Quantized Model | |
| ├── README.md # Model documentation | |
| ``` | |
| ## ⚡ Quantization Details | |
| Post-training quantization was applied using PyTorch's built-in quantization framework. The model was quantized to Float16 (FP16) to reduce model size and improve inference efficiency while balancing accuracy. | |
| ## 📂 Repository Structure | |
| ``` | |
| . | |
| ├── model/ # Contains the quantized model files | |
| ├── tokenizer_config/ # Tokenizer configuration and vocabulary files | |
| ├── model.safetensors/ # Quantized Model | |
| ├── README.md # Model documentation | |
| ``` | |
| ## ⚠️ Limitations | |
| - The model may struggle with highly ambiguous sentences. | |
| - Quantization may lead to slight degradation in accuracy compared to full-precision models. | |
| - Performance may vary across different writing styles and sentence structures. | |
| ## 🤝 Contributing | |
| Contributions are welcome! Feel free to open an issue or submit a pull request if you have suggestions or improvements. | |