Update app.py
Browse files
app.py
CHANGED
@@ -1,36 +1,4 @@
|
|
1 |
-
import
|
2 |
-
import gradio
|
3 |
-
from transformers import pipeline
|
4 |
-
from bs4 import BeautifulSoup
|
5 |
-
import requests
|
6 |
-
|
7 |
-
def SUMMARIZE(Url):
|
8 |
-
summarizer = pipeline("summarization", model="stevhliu/my_awesome_billsum_model")
|
9 |
-
r = requests.get(Url)
|
10 |
-
soup = BeautifulSoup(r.text, 'html.parser')
|
11 |
-
results = soup.find_all(['hl', 'p'])
|
12 |
-
text = [result.text for result in results]
|
13 |
-
Article = ''.join(text)
|
14 |
-
sentences = Article.split(' ')
|
15 |
-
current_chunk = 0
|
16 |
-
chunks = []
|
17 |
-
for sentence in sentences:
|
18 |
-
if len(chunks) == current_chunk + 1:
|
19 |
-
if len(chunks[current_chunk]) + len(sentence.split(' ')) <= max_chunk:
|
20 |
-
chunks[current_chunk].extend(sentence.split(' '))
|
21 |
-
else:
|
22 |
-
current_chunk += 1
|
23 |
-
chunks.append(sentence.split(' '))
|
24 |
-
else:
|
25 |
-
#print(current_chunk)
|
26 |
-
chunks.append(sentence.split(' '))
|
27 |
-
|
28 |
-
for chunk_id in range(len(chunks)):
|
29 |
-
chunks[chunk_id] = ' '.join(chunks[chunk_id])
|
30 |
-
res = summarizer(chunks, max_length=120, min_length=30, do_sample=False)
|
31 |
-
for i in range(len(res)):
|
32 |
-
return res[i].values()
|
33 |
-
|
34 |
|
35 |
interface = gradio.Interface(fn=SUMMARIZE,
|
36 |
inputs=gradio.TextArea(lines=2, value="https://medium.com/analytics-vidhya/openai-gpt-3-language-models-are-few-shot-learners-82531b3d3122"),
|
|
|
1 |
+
from util import SUMMARIZE
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
|
3 |
interface = gradio.Interface(fn=SUMMARIZE,
|
4 |
inputs=gradio.TextArea(lines=2, value="https://medium.com/analytics-vidhya/openai-gpt-3-language-models-are-few-shot-learners-82531b3d3122"),
|