Create README.md
Browse files# Model Card: BERT-based CEFR Classifier
## Overview
This repository contains a model trained to predict Common European Framework of Reference (CEFR) levels for a given text using a BERT-based model architecture. The model was fine-tuned on the CEFR dataset, and the `bert-base-...` pre-trained model was used as the base.
## Model Details
- Model architecture: BERT (base model: `bert-base-...`)
- Task: CEFR level prediction for text classification
- Training dataset: CEFR dataset
- Fine-tuning: Epochs, Loss, Accuracy, etc.
## Performance
The model's performance during training is summarized below:
| Epoch | Training Loss | Validation Loss | Accuracy |
|-------|---------------|-----------------|----------|
| 1 | 1.491800 | 1.319211 | 0.420690 |
| 2 | 1.238600 | 0.864768 | 0.700447 |
| 3 | 0.813200 | 0.538081 | 0.815057 |
Additional metrics:
- Training Loss: 1.1851
- Training Runtime: 7465.51 seconds
- Training Samples per Second: 7.633
- Total Floating Point Operations: 1.499392196785152e+16
## Usage
1. Install the required libraries by running `pip install transformers`.
2. Load the trained model and use it for CEFR-level prediction.
```python
from transformers import pipeline
# Load the model
model = pipeline("text-classification", model="your-model-name-here")
# Predict CEFR levels
result = model("Sample text for CEFR classification.")
print(result)