---
language:
- ku
- kmr
---

# Kurdish Kurmanji BPE Tokenizer

This tokenizer is trained on Kurdish Kurmanji text data using Byte-Pair Encoding (BPE).
It's trained [using](https://huggingface.co/datasets/muzaffercky/kurdish-kurmanji-articles) this dataset

## Details

- **Tokenization Method**: BPE (Byte-Pair Encoding)
- **Vocabulary Size**: 20,000 tokens
- **Special Tokens**: [UNK], [CLS], [SEP], [PAD], [MASK]
- **Language**: Kurdish Kurmanji

## Usage

### Encoding

```python
from transformers import PreTrainedTokenizerFast

tokenizer = PreTrainedTokenizerFast.from_pretrained("muzaffercky/kurdish-kurmanji-tokenizer", revision="v1.0")

text = "Havîn bi gelemperî bi betlane, derya û tavê tê naskirin, lê her kes nikare biçe betlaneyê"

tokens = tokenizer.tokenize(text)

print(tokens)

# Output: ['Hav', 'în bi ', 'gelemperî bi ', 'bet', 'l', 'ane, ', 'der', 'ya û ', 'tavê ', 'tê nas', 'kirin, lê ', 'her kes', ' nikare ', 'biçe ', 'bet', 'lan', 'ey', 'ê']

ids = tokenizer.encode(text)

print(ids)

# Output: [19889, 6494, 19754, 2055, 82, 8227, 337, 3349, 16407, 5790, 11235, 2584, 5479, 5732, 2055, 9479, 287, 140]
```

### Decoding

This tokenizer includes spaces within some tokens (e.g., 'Ez ê ', 'di vê '), which causes the default tokenizer.decode()
method from the transformers library to add extra spaces between tokens during decoding.
To decode text correctly and preserve the original formatting, decode each token ID individually and join the results without spaces.

Use the following code example:

```python
from transformers import PreTrainedTokenizerFast

tokenizer = PreTrainedTokenizerFast.from_pretrained("muzaffercky/kurdish-kurmanji-tokenizer", revision="v1.0")

text = "Havîn bi gelemperî bi betlane, derya û tavê tê naskirin, lê her kes nikare biçe betlaneyê"

ids = tokenizer.encode(text)

individual_tokens = [tokenizer.decode([id]) for id in ids]
decoded_text = "".join(individual_tokens)

print(decoded_text)  # Output: Havîn bi gelemperî bi betlane, derya û tavê tê naskirin, lê her kes nikare biçe betlaneyê
```