metadata
language: de
widget:
- text: My name is Karl and I live in Aachen.
tags:
- translation
datasets:
- wmt19
license: gpl
model-index:
- name: Tanhim/translation-En2De
results:
- task:
type: translation
name: Translation
dataset:
name: wmt19
type: wmt19
config: de-en
split: validation
metrics:
- name: BLEU
type: bleu
value: 43.3134
verified: true
- name: loss
type: loss
value: 0.919737696647644
verified: true
- name: gen_len
type: gen_len
value: 27.8909
verified: true
English to German Translation
Model Name: Tanhim/translation-En2De
language: German or Deutsch
thumbnail: https://huggingface.co/Tanhim/translation-En2De
How to use
You can use this model directly with a pipeline for machine translation. Since the generation relies on some randomness, I set a seed for reproducibility:
>>> from transformers import pipeline, set_seed
>>> text_En2De= pipeline('translation', model='Tanhim/translation-En2De', tokenizer='Tanhim/translation-En2De')
>>> set_seed(42)
>>> text_En2De("My name is Karl and I live in Aachen")