Z

Zeroshot Classification De

Developed by michaelp11
A German zero-shot classification model fine-tuned on DeBERTa-v3-base, suitable for multi-label classification tasks
Downloads 46
Release Time : 2/4/2024

Model Overview

This model is a fine-tuned version on the michaelp11/wiki-tags dataset, specifically designed for German zero-shot classification tasks, capable of classifying text based on provided candidate labels.

Model Features

Zero-shot classification capability
Can classify new categories without task-specific training data
Multi-label support
Supports assigning multiple relevant labels to text simultaneously
High accuracy
Achieves an accuracy and F1 score of 0.859 on the evaluation set

Model Capabilities

German text classification
Multi-label classification
Zero-shot learning

Use Cases

Content classification
Geographic classification
Determines geographic attributes based on text descriptions
Can accurately identify types of locations described in text (e.g., village, city)
Topic classification
Performs topic classification on German text content
Can accurately classify text topics based on provided candidate labels
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase