R

Robbertje 1 Gb Bort

Developed by DTAI-KULeuven
RobBERTje is a series of distilled Dutch BERT models based on RobBERT, offering various model sizes and training configurations.
Downloads 63
Release Time : 3/2/2022

Model Overview

RobBERTje is a series of distilled models based on RobBERT, specifically designed for Dutch and Flemish languages, suitable for various natural language processing tasks.

Model Features

Multiple training versions
Provides models with four different training configurations: non-shuffled, shuffled, merged, and BORT versions
Dutch language optimization
Language model specifically optimized for Dutch and Flemish
Lightweight
Fewer parameters and lower computational resource requirements compared to the original RobBERT model

Model Capabilities

Text understanding
Text generation
Named entity recognition
Part-of-speech tagging
Natural language inference

Use Cases

Text analysis
Sentiment analysis
Analyze sentiment tendencies in Dutch text
Named entity recognition
Identify entities such as person names and locations in Dutch text
Achieved 90.2% accuracy on DBRD dataset (non-shuffled version)
Language understanding
Natural language inference
Determine logical relationships between two Dutch sentences
Achieved 83.4% accuracy on SICK-NL dataset
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase