Sai5480/monolingual-tokenizer-native-mar-vocab-128000

Monolingual Tokenizer - Marathi (Vocab 128000)

This is a monolingual tokenizer trained on Marathi text with vocabulary size 128000.

Usage

from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("monolingual-tokenizer-native-mar-vocab-128000")

Files

• mar.model: SentencePiece model file
• mar.vocab: Vocabulary file
• config.json: Tokenizer configuration

Training Details

• Language: Marathi (mar)
• Vocabulary Size: 128000
• Model Type: SentencePiece Unigram