Spaces:
Runtime error
Runtime error
Create new file
Browse files
model.py
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import nltk
|
2 |
+
import ssl
|
3 |
+
import re
|
4 |
+
try:
|
5 |
+
_create_unverified_https_context = ssl._create_unverified_context
|
6 |
+
except AttributeError:
|
7 |
+
pass
|
8 |
+
else:
|
9 |
+
ssl._create_default_https_context = _create_unverified_https_context
|
10 |
+
nltk.download('punkt')
|
11 |
+
nltk.download('stopwords')
|
12 |
+
from transformers import BartTokenizer, PegasusTokenizer
|
13 |
+
from transformers import BartForConditionalGeneration, PegasusForConditionalGeneration
|
14 |
+
from tqdm.notebook import tqdm
|
15 |
+
|
16 |
+
|
17 |
+
|
18 |
+
class Abstractive_Summarization_Model:
|
19 |
+
def __init__(self):
|
20 |
+
self.text = None
|
21 |
+
self.IS_CNNDM = True # whether to use CNNDM dataset or XSum dataset
|
22 |
+
self.LOWER = False
|
23 |
+
self.max_length = 1024 if self.IS_CNNDM else 512
|
24 |
+
self.model, self.tokenizer = self.load_model()
|
25 |
+
|
26 |
+
def load_model(self):
|
27 |
+
# Load our model checkpoints
|
28 |
+
print('[INFO]: Loading model ...')
|
29 |
+
if self.IS_CNNDM:
|
30 |
+
model = BartForConditionalGeneration.from_pretrained('Yale-LILY/brio-cnndm-uncased')
|
31 |
+
tokenizer = BartTokenizer.from_pretrained('Yale-LILY/brio-cnndm-uncased')
|
32 |
+
else:
|
33 |
+
model = PegasusForConditionalGeneration.from_pretrained('Yale-LILY/brio-xsum-cased')
|
34 |
+
tokenizer = PegasusTokenizer.from_pretrained('Yale-LILY/brio-xsum-cased')
|
35 |
+
print('[INFO]: Model Successfully Loaded :)')
|
36 |
+
return model, tokenizer
|
37 |
+
|
38 |
+
|
39 |
+
def summarize(self, text):
|
40 |
+
# generation example
|
41 |
+
if self.LOWER:
|
42 |
+
article = text.lower()
|
43 |
+
else:
|
44 |
+
article = text
|
45 |
+
inputs = self.tokenizer([article], max_length=self.max_length, return_tensors="pt", truncation=True)
|
46 |
+
# Generate Summary
|
47 |
+
summary_ids = self.model.generate(inputs["input_ids"])
|
48 |
+
|
49 |
+
return self.tokenizer.batch_decode(summary_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
|