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 | |