Update README.md

README.md

@@ -7,10 +7,11 @@ metrics:
 library_name: transformers
 pipeline_tag: text-classification
 ---
-This MistralAI was fined-tuned on nuclear energy data from twitter/X. The classification accuracy obtained is 94%. \
-The number of labels is 3: {0: Negative, 1: Neutral, 2: Positive}
+This MistralAI 7B was fined-tuned on nuclear energy data from twitter/X. The classification accuracy obtained is 94%. \
+The number of labels is 3: {0: Negative, 1: Neutral, 2: Positive} \
+Warning: You need sufficient GPU to run this model.
 
-This is an example to use it
+This is an example to use it, it worked on 8 GB Nvidia-RTX 4060 
 ```bash
 from transformers import AutoTokenizer
 from transformers import pipeline
@@ -29,11 +30,8 @@ if tokenizer.pad_token is None:
 model = AutoModelForSequenceClassification.from_pretrained(checkpoint, 
                                                        num_labels=3,
                                                        id2label=id2label, 
-                                                       label2id=label2id)
-
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-model.to(device)
-
+                                                       label2id=label2id,
+                                                       device_map='auto')
 
 sentiment_task = pipeline("sentiment-analysis", 
                           model=model,