metadata

library_name: transformers
tags:
  - sentiment
license: mit
datasets:
  - scikit-learn/imdb
language:
  - en
metrics:
  - accuracy
base_model:
  - distilbert/distilbert-base-uncased
pipeline_tag: text-classification

DistilBERT Sentiment Classifier (IMDb) — saibapanku/distilbert-sentiment

This is a fine-tuned DistilBERT model for binary sentiment classification trained on the IMDb dataset. The model classifies movie reviews as either positive or negative.

Model Details

Model name: saibapanku/distilbert-sentiment
Base model: distilbert-base-uncased
Task: Sequence Classification (Sentiment Analysis)
Dataset: IMDb
Labels:
- 0: Negative
- 1: Positive

How to Use

You can load and use the model directly with 🤗 Transformers:

from transformers import pipeline

classifier = pipeline("text-classification", model="saibapanku/distilbert-sentiment")
print(classifier("This movie was absolutely amazing!"))

Training Configuration

Training method: Hugging Face Trainer
Epochs: 3
Batch size: 16
Max sequence length: 256 tokens
Learning rate: default
Weight decay: 0.01
Evaluation strategy: per epoch
Metric used: Accuracy
Subset used: 2,000 train / 1,000 test samples (for demo purposes)

Example Output: [{'label': 'positive', 'score': 0.9843}]

Limitations

This model was trained on a small subset of the IMDb dataset and may not generalize well to all types of reviews.

Performance on domain-specific or multi-lingual content is not guaranteed.

License

This model is distributed under the MIT License.

Feel free to fine-tune further or adapt it for your specific sentiment analysis tasks!