Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
You need a custom version of the `tokenizers` library to use this tokenizer.
|
| 2 |
+
|
| 3 |
+
To install this custom version you can:
|
| 4 |
+
```bash
|
| 5 |
+
pip install transformers
|
| 6 |
+
git clone https://github.com/huggingface/tokenizers.git
|
| 7 |
+
cd tokenizers
|
| 8 |
+
git checkout bigscience_fork
|
| 9 |
+
cd bindings/python
|
| 10 |
+
pip install setuptools_rust
|
| 11 |
+
pip install -e .
|
| 12 |
+
```
|
| 13 |
+
|
| 14 |
+
and then to load it, do:
|
| 15 |
+
```python
|
| 16 |
+
from transformers import AutoTokenizer
|
| 17 |
+
|
| 18 |
+
tokenizer = AutoTokenizer.from_pretrained("bigscience-catalogue-data-dev/byte-level-bpe-tokenizer-no-norm-250k-whitespace-and-eos-regex-alpha-v3-dedup-lines-articles")
|
| 19 |
+
```
|