julien-c HF staff commited on
Commit
8d27c3e
1 Parent(s): 4988b91

Migrate model card from transformers-repo

Browse files

Read announcement at https://discuss.huggingface.co/t/announcement-all-model-cards-will-be-migrated-to-hf-co-model-repos/2755
Original file history: https://github.com/huggingface/transformers/commits/master/model_cards/allegro/herbert-klej-cased-tokenizer-v1/README.md

Files changed (1) hide show
  1. README.md +44 -0
README.md ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: pl
3
+ ---
4
+
5
+ # HerBERT tokenizer
6
+
7
+ **[HerBERT](https://en.wikipedia.org/wiki/Zbigniew_Herbert)** tokenizer is a character level byte-pair encoding with
8
+ vocabulary size of 50k tokens. The tokenizer was trained on [Wolne Lektury](https://wolnelektury.pl/) and a publicly available subset of
9
+ [National Corpus of Polish](http://nkjp.pl/index.php?page=14&lang=0) with [fastBPE](https://github.com/glample/fastBPE) library.
10
+ Tokenizer utilize `XLMTokenizer` implementation from [transformers](https://github.com/huggingface/transformers).
11
+
12
+ ## Tokenizer usage
13
+ Herbert tokenizer should be used together with [HerBERT model](https://huggingface.co/allegro/herbert-klej-cased-v1):
14
+ ```python
15
+ from transformers import XLMTokenizer, RobertaModel
16
+
17
+ tokenizer = XLMTokenizer.from_pretrained("allegro/herbert-klej-cased-tokenizer-v1")
18
+ model = RobertaModel.from_pretrained("allegro/herbert-klej-cased-v1")
19
+
20
+ encoded_input = tokenizer.encode("Kto ma lepszą sztukę, ma lepszy rząd – to jasne.", return_tensors='pt')
21
+ outputs = model(encoded_input)
22
+ ```
23
+
24
+ ## License
25
+ CC BY-SA 4.0
26
+
27
+ ## Citation
28
+ If you use this tokenizer, please cite the following paper:
29
+ ```
30
+ @misc{rybak2020klej,
31
+ title={KLEJ: Comprehensive Benchmark for Polish Language Understanding},
32
+ author={Piotr Rybak and Robert Mroczkowski and Janusz Tracz and Ireneusz Gawlik},
33
+ year={2020},
34
+ eprint={2005.00630},
35
+ archivePrefix={arXiv},
36
+ primaryClass={cs.CL}
37
+ }
38
+ ```
39
+ Paper is accepted at ACL 2020, as soon as proceedings appear, we will update the BibTeX.
40
+
41
+ ## Authors
42
+ Tokenizer was created by **Allegro Machine Learning Research** team.
43
+
44
+ You can contact us at: <a href="mailto:klejbenchmark@allegro.pl">klejbenchmark@allegro.pl</a>