Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
datasets:
|
| 3 |
+
- IteraTeR_full_sent
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# IteraTeR BART model
|
| 7 |
+
This model was obtained by fine-tuning [facebook/bart-base](https://huggingface.co/facebook/bart-base) on [IteraTeR-full-sent](https://huggingface.co/datasets/wanyu/IteraTeR_full_sent) dataset.
|
| 8 |
+
|
| 9 |
+
Paper: [Understanding Iterative Revision from Human-Written Text](https://arxiv.org/abs/2203.03802) <br>
|
| 10 |
+
Authors: Wanyu Du, Vipul Raheja, Dhruv Kumar, Zae Myung Kim, Melissa Lopez, Dongyeop Kang
|
| 11 |
+
|
| 12 |
+
## Usage
|
| 13 |
+
```python
|
| 14 |
+
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
|
| 15 |
+
|
| 16 |
+
tokenizer = AutoTokenizer.from_pretrained("wanyu/IteraTeR-BART")
|
| 17 |
+
model = AutoModelForSeq2SeqLM.from_pretrained("wanyu/IteraTeR-BART")
|
| 18 |
+
before_input = '<fluency> I likes coffee.'
|
| 19 |
+
model_input = tokenizer(before_input, return_tensors='pt')
|
| 20 |
+
model_outputs = model.generate(**model_input, num_beams=8, max_length=1024)
|
| 21 |
+
after_text = tokenizer.batch_decode(model_outputs, skip_special_tokens=True)[0]
|
| 22 |
+
```
|