File size: 1,095 Bytes
19f0189 834e7e0 19f0189 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 |
---
language: en
license: mit
tags:
- GECToR_gotutiyan
- grammatical error correction
---
# gector sample
This is an unofficial pretrained model of GECToR ([Omelianchuk+ 2020](https://aclanthology.org/2020.bea-1.16/)).
### How to use
The code is avaliable from https://github.com/gotutiyan/gector.
CLI
```sh
python predict.py --input <raw text file> --restore_dir gotutiyan/gector-deberta-large-5k --out <path to output file>
```
API
```py
from transformers import AutoTokenizer
from gector.modeling import GECToR
from gector.predict import predict, load_verb_dict
import torch
model_id = 'gotutiyan/gector-deberta-large-5k'
model = GECToR.from_pretrained(model_id)
if torch.cuda.is_available():
model.cuda()
tokenizer = AutoTokenizer.from_pretrained(model_id)
encode, decode = load_verb_dict('data/verb-form-vocab.txt')
srcs = [
'This is a correct sentence.',
'This are a wrong sentences'
]
corrected = predict(
model, tokenizer, srcs,
encode, decode,
keep_confidence=0.0,
min_error_prob=0.0,
n_iteration=5,
batch_size=2,
)
print(corrected)
```
|