Commit
•
d82eb4f
1
Parent(s):
95c7162
uploaded readme
Browse files
README.md
ADDED
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Quantization made by Richard Erkhov.
|
2 |
+
|
3 |
+
[Github](https://github.com/RichardErkhov)
|
4 |
+
|
5 |
+
[Discord](https://discord.gg/pvy7H8DZMG)
|
6 |
+
|
7 |
+
[Request more models](https://github.com/RichardErkhov/quant_request)
|
8 |
+
|
9 |
+
|
10 |
+
gpt2-small-italian - bnb 4bits
|
11 |
+
- Model creator: https://huggingface.co/GroNLP/
|
12 |
+
- Original model: https://huggingface.co/GroNLP/gpt2-small-italian/
|
13 |
+
|
14 |
+
|
15 |
+
|
16 |
+
|
17 |
+
Original model description:
|
18 |
+
---
|
19 |
+
language: it
|
20 |
+
tags:
|
21 |
+
- adaption
|
22 |
+
- recycled
|
23 |
+
- gpt2-small
|
24 |
+
pipeline_tag: text-generation
|
25 |
+
---
|
26 |
+
|
27 |
+
# GPT-2 recycled for Italian (small)
|
28 |
+
[Wietse de Vries](https://www.semanticscholar.org/author/Wietse-de-Vries/144611157) •
|
29 |
+
[Malvina Nissim](https://www.semanticscholar.org/author/M.-Nissim/2742475)
|
30 |
+
|
31 |
+
## Model description
|
32 |
+
|
33 |
+
This model is based on the small OpenAI GPT-2 ([`gpt2`](https://huggingface.co/gpt2)) model.
|
34 |
+
|
35 |
+
For details, check out our paper on [arXiv](https://arxiv.org/abs/2012.05628) and the code on [Github](https://github.com/wietsedv/gpt2-recycle).
|
36 |
+
|
37 |
+
|
38 |
+
## Related models
|
39 |
+
|
40 |
+
### Dutch
|
41 |
+
- [`gpt2-small-dutch-embeddings`](https://huggingface.co/GroNLP/gpt2-small-dutch-embeddings): Small model size with only retrained lexical embeddings.
|
42 |
+
- [`gpt2-small-dutch`](https://huggingface.co/GroNLP/gpt2-small-dutch): Small model size with retrained lexical embeddings and additional fine-tuning of the full model. (**Recommended**)
|
43 |
+
- [`gpt2-medium-dutch-embeddings`](https://huggingface.co/GroNLP/gpt2-medium-dutch-embeddings): Medium model size with only retrained lexical embeddings.
|
44 |
+
|
45 |
+
### Italian
|
46 |
+
- [`gpt2-small-italian-embeddings`](https://huggingface.co/GroNLP/gpt2-small-italian-embeddings): Small model size with only retrained lexical embeddings.
|
47 |
+
- [`gpt2-small-italian`](https://huggingface.co/GroNLP/gpt2-small-italian): Small model size with retrained lexical embeddings and additional fine-tuning of the full model. (**Recommended**)
|
48 |
+
- [`gpt2-medium-italian-embeddings`](https://huggingface.co/GroNLP/gpt2-medium-italian-embeddings): Medium model size with only retrained lexical embeddings.
|
49 |
+
|
50 |
+
|
51 |
+
## How to use
|
52 |
+
|
53 |
+
```python
|
54 |
+
from transformers import pipeline
|
55 |
+
|
56 |
+
pipe = pipeline("text-generation", model="GroNLP/gpt2-small-italian")
|
57 |
+
```
|
58 |
+
|
59 |
+
```python
|
60 |
+
from transformers import AutoTokenizer, AutoModel, TFAutoModel
|
61 |
+
|
62 |
+
tokenizer = AutoTokenizer.from_pretrained("GroNLP/gpt2-small-italian")
|
63 |
+
model = AutoModel.from_pretrained("GroNLP/gpt2-small-italian") # PyTorch
|
64 |
+
model = TFAutoModel.from_pretrained("GroNLP/gpt2-small-italian") # Tensorflow
|
65 |
+
```
|
66 |
+
|
67 |
+
## BibTeX entry
|
68 |
+
|
69 |
+
```bibtex
|
70 |
+
@misc{devries2020good,
|
71 |
+
title={As good as new. How to successfully recycle English GPT-2 to make models for other languages},
|
72 |
+
author={Wietse de Vries and Malvina Nissim},
|
73 |
+
year={2020},
|
74 |
+
eprint={2012.05628},
|
75 |
+
archivePrefix={arXiv},
|
76 |
+
primaryClass={cs.CL}
|
77 |
+
}
|
78 |
+
```
|
79 |
+
|
80 |
+
|