🤖 DistilBERT Fine-Tuned on IMDb Movie Reviews for Sentiment Classification
DistilBERT model fine-tuned on the IMDb dataset for binary sentiment classification (Positive/Negative).
🧠 Model Description
- Architecture: DistilBERT (lighter version of BERT)
- Task: Sentiment Analysis / Text Classification
- Dataset: IMDb movie reviews (binary sentiment: positive or negative)
- Base Model:
distilbert-base-uncased
- Fine-tuned using: Hugging Face Transformers + Trainer API
📊 Training Details
- Training Samples: 5,000 (subset of IMDb train split)
- Test Samples: 1,000 (subset of IMDb test split)
- Epochs: 2
- Batch Size: 8
- Optimizer: AdamW
- Evaluation Metric: Accuracy
🚀 How to Use
You can directly use this model for inference using transformers pipeline:
from transformers import pipeline
classifier = pipeline("sentiment-analysis", model="your-username/distilbert-finetuned-imdb-sentiment")
print(classifier("This movie was absolutely amazing!"))
# Output: [{'label': 'POSITIVE', 'score': 0.99}]
