Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- ru
|
4 |
+
- en
|
5 |
+
- ru-RU
|
6 |
+
tags:
|
7 |
+
- xlm-roberta-large
|
8 |
+
datasets:
|
9 |
+
- IlyaGusev/headline_cause
|
10 |
+
license: apache-2.0
|
11 |
+
---
|
12 |
+
|
13 |
+
# XLM-RoBERTa HeadlineCause Simple
|
14 |
+
|
15 |
+
## Model description
|
16 |
+
|
17 |
+
TBA
|
18 |
+
|
19 |
+
## Intended uses & limitations
|
20 |
+
|
21 |
+
#### How to use
|
22 |
+
|
23 |
+
```python
|
24 |
+
from tqdm.notebook import tqdm
|
25 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
|
26 |
+
|
27 |
+
def get_batch(data, batch_size):
|
28 |
+
start_index = 0
|
29 |
+
while start_index < len(data):
|
30 |
+
end_index = start_index + batch_size
|
31 |
+
batch = data[start_index:end_index]
|
32 |
+
yield batch
|
33 |
+
start_index = end_index
|
34 |
+
|
35 |
+
|
36 |
+
def pipe_predict(data, pipe, batch_size=64):
|
37 |
+
raw_preds = []
|
38 |
+
for batch in tqdm(get_batch(data, batch_size)):
|
39 |
+
raw_preds += pipe(batch)
|
40 |
+
return raw_preds
|
41 |
+
|
42 |
+
MODEL_NAME = TOKENIZER_NAME = "IlyaGusev/xlm_roberta_large_headline_cause_simple"
|
43 |
+
tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_NAME, do_lower_case=False)
|
44 |
+
model = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME)
|
45 |
+
model.eval()
|
46 |
+
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer, framework="pt", return_all_scores=True)
|
47 |
+
texts = [
|
48 |
+
(
|
49 |
+
"Judge issues order to allow indoor worship in NC churches",
|
50 |
+
"Some local churches resume indoor services after judge lifted NC governor’s restriction"
|
51 |
+
),
|
52 |
+
(
|
53 |
+
"Gov. Kevin Stitt defends $2 million purchase of malaria drug touted by Trump",
|
54 |
+
"Oklahoma spent $2 million on malaria drug touted by Trump"
|
55 |
+
),
|
56 |
+
(
|
57 |
+
"Песков опроверг свой перевод на удаленку",
|
58 |
+
"Дмитрий Песков перешел на удаленку"
|
59 |
+
)
|
60 |
+
]
|
61 |
+
pipe_predict(texts, pipe)
|
62 |
+
```
|
63 |
+
|
64 |
+
#### Limitations and bias
|
65 |
+
|
66 |
+
[More Information Needed]
|
67 |
+
|
68 |
+
## Training data
|
69 |
+
|
70 |
+
[More Information Needed]
|
71 |
+
|
72 |
+
## Training procedure
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Eval results
|
77 |
+
|
78 |
+
[More Information Needed]
|
79 |
+
|
80 |
+
### BibTeX entry and citation info
|
81 |
+
|
82 |
+
```bibtex
|
83 |
+
@misc{gusev2021headlinecause,
|
84 |
+
title={HeadlineCause: A Dataset of News Headlines for Detecting Casualties},
|
85 |
+
author={Ilya Gusev and Alexey Tikhonov},
|
86 |
+
year={2021},
|
87 |
+
eprint={2108.12626},
|
88 |
+
archivePrefix={arXiv},
|
89 |
+
primaryClass={cs.CL}
|
90 |
+
}
|
91 |
+
```
|