Update README.md

README.md

@@ -99,6 +99,19 @@ def compute_loss(self, model, inputs, return_outputs=False, num_items_in_batch=N
         return (loss, outputs) if return_outputs else loss
 ```
 
+# Usage
+
+```python
+from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
+model = pipeline(task='sentiment-analysis', model='alexander-sh/mDeBERTa-v3-multi-sent', device='cuda')
+model('Keep your face always toward the sunshine—and shadows will fall behind you.')
+>>> [{'label': 'positive', 'score': 0.6478521227836609}]
+model('I am not coming with you.')
+>>> [{'label': 'neutral', 'score': 0.790919840335846}]
+model("I am hating that my transformer model don't work properly.")
+>>> [{'label': 'negative', 'score': 0.7474458813667297}]
+```
+
 # Evaluation and comparison with Vanilla and GPT-4o model:
 
 | Dataset          | Model  | F1     | Accuracy |