metadata
datasets:
- sungmogi/en2ko_hiphop
language:
- en
- ko
tags:
- small100
pipeline_tag: translation
inference:
parameters:
src_lang: en
tgt_lang: ko
Model Description
en2ko_hiphop_small-100 is a fine-tuned version of SMaLL-100 on en2ko_hiphop dataset.
How to use
Here is how to use this model to translate English text to Korean text using Transformers Pipeline:
from transformers import pipeline
pipe = pipeline("translation", model="sungmogi/en2ko_hiphop_small-100", src_lang="en", tgt_lang="ko")
pipe(input_text)
Training Hyperparameters
- per_device_train_batch_size: 4
- per_device_eval_batch_size: 4
- weight_decay: 0.01
- num_train_epochs: 4
- num_devices: 4
- learning_rate: 4e-5