Update README.md
Browse files
README.md
CHANGED
|
@@ -74,6 +74,9 @@ The dataset is balanced across the six emotion classes, promoting effective lear
|
|
| 74 |
- **Tokenization**: Text was tokenized using the DistilBERT tokenizer, with a maximum sequence length of 512 tokens. Padding and truncation ensured uniform input sizes.
|
| 75 |
- **Data Formatting**: Converted to PyTorch tensors for training compatibility.
|
| 76 |
|
|
|
|
|
|
|
|
|
|
| 77 |
### Training Hyperparameters
|
| 78 |
Fine-tuning was conducted using the Hugging Face `Trainer` API with:
|
| 79 |
- **Epochs**: 3
|
|
|
|
| 74 |
- **Tokenization**: Text was tokenized using the DistilBERT tokenizer, with a maximum sequence length of 512 tokens. Padding and truncation ensured uniform input sizes.
|
| 75 |
- **Data Formatting**: Converted to PyTorch tensors for training compatibility.
|
| 76 |
|
| 77 |
+
## Demo
|
| 78 |
+
Try the model in action [here](https://huggingface.co/spaces/YonasMersha/emotion-classifier).
|
| 79 |
+
|
| 80 |
### Training Hyperparameters
|
| 81 |
Fine-tuning was conducted using the Hugging Face `Trainer` API with:
|
| 82 |
- **Epochs**: 3
|