sentiment-tutorial / README.md
Rameen191's picture
Upload README.md with huggingface_hub
e63073a verified
|
Raw
History Blame Contribute Delete
817 Bytes
metadata
language: en
license: apache-2.0
tags:
  - sentiment-analysis
  - text-classification
  - distilbert
datasets:
  - imdb
metrics:
  - accuracy
  - f1
pipeline_tag: text-classification

sentiment-tutorial

Fine-tuned distilbert-base-uncased for binary sentiment classification.

Intended Use

Classify English text as positive or negative.

Training Procedure

  • Base model: distilbert-base-uncased
  • Epochs: 2
  • Learning rate: 2e-5
  • Batch size: 32
  • Max length: 128

Evaluation Results

Accuracy: 0.869

Precision: 0.878

Recall: 0.858

F1: 0.868

Limitations

  • Binary classification only
  • English only
  • Movie reviews domain

Usage

from transformers import pipeline

classifier = pipeline( "sentiment-analysis", model="Rameen191/sentiment-tutorial" )

classifier("This was a great experience!")