distilbert-imdb-sentiment

DistilBERT IMDB Sentiment Classifier

A fine-tuned DistilBERT model for binary sentiment analysis on the IMDB movie reviews dataset.

• Task: Text Classification (Sentiment Analysis)
• Labels: POSITIVE / NEGATIVE
• Base model: distilbert-base-uncased
• Framework: 🤗 Transformers + PyTorch

---

Model Description

This model is fine-tuned on a subset of the IMDB Reviews dataset to classify movie reviews as positive or negative. It is designed as a clean, end-to-end NLP project demonstrating:

• Dataset loading
• Tokenization
• Fine-tuning with Trainer
• Evaluation with Accuracy and F1
• Inference using pipeline

---

Training Details

• Dataset: IMDB (subset)
• Train samples: 4,000
• Validation samples: 1,000
• Epochs: 3
• Optimizer: AdamW
• Batch size (train): 8
• Batch size (eval): 16
• Max sequence length: 512

---

Evaluation Results

Note: Results are reported on a validation subset and are intended for demonstration purposes.

---

Usage

Quick Inference

from transformers import pipeline

classifier = pipeline(
    "text-classification",
    model="qusai00/distilbert-imdb-sentiment"
)

print(classifier("This movie was amazing, I loved it!"))
print(classifier("Terrible film, waste of time."))


Files in this Repository

. model.safetensors — fine-tuned model weights

. config.json — model configuration

. tokenizer.json / vocab.txt — tokenizer files

. training_args.bin — training configuration


Author

Qusai (qusai00)
AI / NLP Engineer in training