cahya/distilbert-base-indonesian

The Indonesian language model is a distilled version of the Indonesian BERT base model . It is uncased . It was pre-trained with 522 of 522 Indonesian datasets and 1GB of Indonesian newspapers . The model was distiled with a vocabulary of 32,000 words . It can be used for text classification, text generation and other downstream tasks . It has been used in the form of the Sentence Sentence A [SEP] Sentence BERT (SSEP) The model is available at Transformer based Indonesian Language Models. It is available to use in PyTorch and in Tensorflow. It can also be used in other language models that have been trained with Indonesian,