| from transformers import AutoTokenizer, AutoModelForSequenceClassification | |
| tokenizer = AutoTokenizer.from_pretrained(model_checkpoint) | |
| model = AutoModelForSequenceClassification.from_pretrained('techthiyanes/Bert_Bahasa_Sentiment') | |
| inputs = tokenizer("saya tidak", return_tensors="pt") | |
| labels = torch.tensor([1]).unsqueeze(0) | |
| outputs = model(**inputs, labels=labels) | |
| loss = outputs.loss | |
| logits = outputs.logits | |
| outputs | |
| hello | |