Spaces:
Runtime error
Runtime error
cache load model funcs for faster load times
Browse files
app.py
CHANGED
@@ -3,13 +3,12 @@ import streamlit as st
|
|
3 |
from extractive_summarizer.model_processors import Summarizer
|
4 |
from transformers import T5Tokenizer, T5ForConditionalGeneration, T5Config
|
5 |
|
6 |
-
def abstractive_summarizer(text : str):
|
7 |
-
abs_model = T5ForConditionalGeneration.from_pretrained('t5-large')
|
8 |
tokenizer = T5Tokenizer.from_pretrained('t5-large')
|
9 |
device = torch.device('cpu')
|
10 |
preprocess_text = text.strip().replace("\n", "")
|
11 |
t5_prepared_text = "summarize: " + preprocess_text
|
12 |
-
tokenized_text = tokenizer.encode(t5_prepared_text, return_tensors="pt").to(
|
13 |
|
14 |
# summmarize
|
15 |
summary_ids = abs_model.generate(tokenized_text,
|
@@ -22,6 +21,17 @@ def abstractive_summarizer(text : str):
|
|
22 |
|
23 |
return abs_summarized_text
|
24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
25 |
if __name__ == "__main__":
|
26 |
# ---------------------------------
|
27 |
# Main Application
|
@@ -42,12 +52,12 @@ if __name__ == "__main__":
|
|
42 |
if summarize_type == "Extractive":
|
43 |
# extractive summarizer
|
44 |
|
45 |
-
ext_model =
|
46 |
summarized_text = ext_model(inp_text, num_sentences=5)
|
47 |
|
48 |
elif summarize_type == "Abstractive":
|
49 |
-
|
50 |
-
summarized_text = abstractive_summarizer(inp_text)
|
51 |
|
52 |
# final summarized output
|
53 |
st.subheader("Summarized text")
|
|
|
3 |
from extractive_summarizer.model_processors import Summarizer
|
4 |
from transformers import T5Tokenizer, T5ForConditionalGeneration, T5Config
|
5 |
|
6 |
+
def abstractive_summarizer(text : str, model):
|
|
|
7 |
tokenizer = T5Tokenizer.from_pretrained('t5-large')
|
8 |
device = torch.device('cpu')
|
9 |
preprocess_text = text.strip().replace("\n", "")
|
10 |
t5_prepared_text = "summarize: " + preprocess_text
|
11 |
+
tokenized_text = tokenizer.encode(t5_prepared_text, return_tensors="pt").to(device)
|
12 |
|
13 |
# summmarize
|
14 |
summary_ids = abs_model.generate(tokenized_text,
|
|
|
21 |
|
22 |
return abs_summarized_text
|
23 |
|
24 |
+
@st.cache()
|
25 |
+
def load_ext_model():
|
26 |
+
model = Summarizer()
|
27 |
+
return model
|
28 |
+
|
29 |
+
@st.cache()
|
30 |
+
def load_abs_model():
|
31 |
+
model = T5ForConditionalGeneration.from_pretrained('t5-large')
|
32 |
+
return model
|
33 |
+
|
34 |
+
|
35 |
if __name__ == "__main__":
|
36 |
# ---------------------------------
|
37 |
# Main Application
|
|
|
52 |
if summarize_type == "Extractive":
|
53 |
# extractive summarizer
|
54 |
|
55 |
+
ext_model = load_ext_model()
|
56 |
summarized_text = ext_model(inp_text, num_sentences=5)
|
57 |
|
58 |
elif summarize_type == "Abstractive":
|
59 |
+
abs_model = load_abs_model()
|
60 |
+
summarized_text = abstractive_summarizer(inp_text, model=abs_model)
|
61 |
|
62 |
# final summarized output
|
63 |
st.subheader("Summarized text")
|