Views
No views yet
bert-base-uncased on the Stanford IMDB dataset for binary sentiment classification.| Parameter | Value |
|---|---|
| Base model | bert-base-uncased |
| Learning rate | 2e-5 |
| Batch size | 4 |
| Epochs | 2 |
| Max sequence length | 512 |
1from transformers import BertForSequenceClassification, BertTokenizer
2
3tokenizer = BertTokenizer.from_pretrained("COMP6713bert/imdb-bert-sentiment")
4model = BertForSequenceClassification.from_pretrained("COMP6713bert/imdb-bert-sentiment")
5
6inputs = tokenizer("This movie was great!", return_tensors="pt", truncation=True, max_length=512)
7with torch.no_grad():
8 outputs = model(**inputs)
9 predicted = torch.argmax(outputs.logits, dim=-1).item()
10
11print("Positive" if predicted == 1 else "Negative")