| language: en | |
| license: apache-2.0 | |
| base_model: bert-base-uncased | |
| tags: | |
| - sentiment-analysis | |
| - imdb | |
| - text-classification | |
| widget: | |
| - text: "This movie was absolutely fantastic! Best film I've seen all year." | |
| # BERT Fine-tuned on IMDB for Sentiment Analysis | |
| Fine-tuned from `bert-base-uncased` on the Stanford IMDB dataset for binary sentiment classification. | |
| ## Training Details | |
| | Parameter | Value | | |
| |-----------|-------| | |
| | Base model | bert-base-uncased | | |
| | Learning rate | 2e-5 | | |
| | Batch size | 4 | | |
| | Epochs | 2 | | |
| | Max sequence length | 512 | | |
| ## Usage | |
| ```python | |
| from transformers import BertForSequenceClassification, BertTokenizer | |
| tokenizer = BertTokenizer.from_pretrained("COMP6713bert/imdb-bert-sentiment") | |
| model = BertForSequenceClassification.from_pretrained("COMP6713bert/imdb-bert-sentiment") | |
| inputs = tokenizer("This movie was great!", return_tensors="pt", truncation=True, max_length=512) | |
| with torch.no_grad(): | |
| outputs = model(**inputs) | |
| predicted = torch.argmax(outputs.logits, dim=-1).item() | |
| print("Positive" if predicted == 1 else "Negative") | |
| ``` | |
| ## Labels | |
| - 0: Negative | |
| - 1: Positive |