Migrate model card from transformers-repo
Browse filesRead 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/google/roberta2roberta_L-24_bbc/README.md
README.md
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
datasets:
|
| 5 |
+
- xsum
|
| 6 |
+
tags:
|
| 7 |
+
- summarization
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# Roberta2Roberta_L-24_bbc EncoderDecoder model
|
| 11 |
+
|
| 12 |
+
The model was introduced in
|
| 13 |
+
[this paper](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn and first released in [this repository](https://tfhub.dev/google/bertseq2seq/roberta24_bbc/1).
|
| 14 |
+
|
| 15 |
+
The model is an encoder-decoder model that was initialized on the `roberta-large` checkpoints for both the encoder
|
| 16 |
+
and decoder and fine-tuned on extreme summarization on the BBC XSum dataset, which is linked above.
|
| 17 |
+
|
| 18 |
+
Disclaimer: The model card has been written by the Hugging Face team.
|
| 19 |
+
|
| 20 |
+
## How to use
|
| 21 |
+
|
| 22 |
+
You can use this model for extreme summarization, *e.g.*
|
| 23 |
+
|
| 24 |
+
```python
|
| 25 |
+
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
|
| 26 |
+
|
| 27 |
+
tokenizer = AutoTokenizer.from_pretrained("google/roberta2roberta_L-24_bbc")
|
| 28 |
+
model = AutoModelForSeq2SeqLM.from_pretrained("google/roberta2roberta_L-24_bbc")
|
| 29 |
+
|
| 30 |
+
article = """The problem is affecting people using the older
|
| 31 |
+
versions of the PlayStation 3, called the "Fat"
|
| 32 |
+
model.The problem isn't affecting the newer PS3
|
| 33 |
+
Slim systems that have been on sale since
|
| 34 |
+
September last year.Sony have also said they are
|
| 35 |
+
aiming to have the problem fixed shortly but is
|
| 36 |
+
advising some users to avoid using their console
|
| 37 |
+
for the time being."We hope to resolve this
|
| 38 |
+
problem within the next 24 hours," a statement
|
| 39 |
+
reads. "In the meantime, if you have a model other
|
| 40 |
+
than the new slim PS3, we advise that you do not
|
| 41 |
+
use your PS3 system, as doing so may result in
|
| 42 |
+
errors in some functionality, such as recording
|
| 43 |
+
obtained trophies, and not being able to restore
|
| 44 |
+
certain data."We believe we have identified that
|
| 45 |
+
this problem is being caused by a bug in the clock
|
| 46 |
+
functionality incorporated in the system."The
|
| 47 |
+
PlayStation Network is used by millions of people
|
| 48 |
+
around the world.It allows users to play their
|
| 49 |
+
friends at games like Fifa over the internet and
|
| 50 |
+
also do things like download software or visit
|
| 51 |
+
online stores."""
|
| 52 |
+
|
| 53 |
+
input_ids = tokenizer(article, return_tensors="pt").input_ids
|
| 54 |
+
output_ids = model.generate(input_ids)[0]
|
| 55 |
+
print(tokenizer.decode(output_ids, skip_special_tokens=True))
|
| 56 |
+
# should output
|
| 57 |
+
# Some Sony PlayStation gamers are being advised to stay away from the network because of a problem with the PlayStation 3 network.
|
| 58 |
+
```
|