Downloads · 30 days
7
1% of all-time downloads
nlpconnect/distilbert-base-cased-wikiproperties-classifier
distilbert-base-cased-wikiproperties-classifier is a text classification model from nlpconnect. Use it when you need a label for a piece of text. It is set up for transformers.
{'author': 'P50', 'capital': 'P36', 'child': 'P40', 'country': 'P17', 'country of origin': 'P495', 'creator': 'P170', 'educated at': 'P69', 'founder': 'P112', 'genre': 'P136', 'headquarters location': 'P159', 'languag…
Downloads · 30 days
7
1% of all-time downloads
All-time downloads
617
Public
Repo size
527 MB
Likes
2
Public
Click a slice to open those files.
.h5263 MB · 100%
From the Hugging Face model README
{'author': 'P50', 'capital': 'P36', 'child': 'P40', 'country': 'P17', 'country of origin': 'P495', 'creator': 'P170', 'educated at': 'P69', 'founder': 'P112', 'genre': 'P136', 'headquarters location': 'P159', 'language of work or name': 'P407', 'located in the administrative territorial entity': 'P131', 'location': 'P276', 'location of formation': 'P740', 'manufacturer': 'P176', 'notable work': 'P800', 'occupation': 'P106', 'owned by': 'P127', 'performer': 'P175', 'place of birth': 'P19', 'place of death': 'P20', 'position played on team / speciality': 'P413', 'record label': 'P264', 'spouse': 'P26'}
1250/1250 [==============================] - 159s 116ms/step - loss: 0.2096 - accuracy: 0.9570 - val_loss: 0.0407 - val_accuracy: 0.9805
from transformers import pipeline
classfier = pipeline("text-classification", model="nlpconnect/distilbert-base-cased-wikiproperties-classifier")