What is this?

BERT classification model for short customer reviews written in Danish.

The model uses 5 classes ranging from 1-5 stars:

  • ⭐ (very poor)
  • ⭐⭐ (poor)
  • ⭐⭐⭐ (neutral)
  • ⭐⭐⭐⭐ (good)
  • ⭐⭐⭐⭐⭐ (very good)

The model is fine-tuned using the pre-trained Danish BERT model.

How to use

Test the model using the 🤗Transformers library pipeline:

from transformers import pipeline

classifier = pipeline("sentiment-analysis", model="KennethTM/danish-bert-review-sentiment")
classifier("Intet virkede og ingen hjælp at hente.")

#[{'label': '⭐', 'score': 0.4953940808773041}]

Or load it using the Auto* classes:

from transformers import AutoTokenizer, AutoModelForSequenceClassification

model = AutoModelForSequenceClassification.from_pretrained("KennethTM/danish-bert-review-sentiment")
tokenizer = AutoTokenizer.from_pretrained("KennethTM/danish-bert-review-sentiment")
Downloads last month
57
Safetensors
Model size
111M params
Tensor type
I64
·
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.