DistilBERT Sentiment Classifier

Fine-tuned distilbert-base-uncased for 2 epochs on the official rotten_tomatoes splits. Labels are NEGATIVE (0) and POSITIVE (1).

Test result

Metric Value
Accuracy 0.8499

Usage

from transformers import pipeline
classifier = pipeline("sentiment-analysis", model="Wadden/distilbert-sentiment-demo")
classifier("This movie was fantastic!")

Interactive demo

The quantized ONNX export is executed directly in the browser with Transformers.js:

Open the interactive Space

Downloads last month
63
Safetensors
Model size
67M params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Dataset used to train Wadden/distilbert-sentiment-demo

Space using Wadden/distilbert-sentiment-demo 1