Tylerus/bert-imdb-sentiment

BERT for IMDb Sentiment Analysis

This is a fine-tuned bert-base-uncased model for binary sentiment classification on the IMDb movie review dataset. It predicts whether a review expresses positive or negative sentiment.

Model Architecture

• Base model: BERT (bert-base-uncased)
• Fine-tuned using: AutoModelForSequenceClassification
• Number of output labels: 2 (positive, negative)

Training Details

• Dataset: IMDb Large Movie Review Dataset
• Epochs: 2
• Optimizer: AdamW
• Evaluation metric: Accuracy
• Final validation accuracy: ~91.6%

How to Use

from transformers import AutoModelForSequenceClassification, AutoTokenizer

model = AutoModelForSequenceClassification.from_pretrained("Tylerus/bert-imdb-sentiment")
tokenizer = AutoTokenizer.from_pretrained("Tylerus/bert-imdb-sentiment")

inputs = tokenizer("This movie was fantastic!", return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits