Upload app.py
Browse files
app.py
ADDED
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
from transformers import T5TokenizerFast, T5ForConditionalGeneration
|
3 |
+
import nltk
|
4 |
+
import math
|
5 |
+
import torch
|
6 |
+
|
7 |
+
model_name = "abokbot/t5-end2end-questions-generation"
|
8 |
+
max_input_length = 512
|
9 |
+
|
10 |
+
st.header("Generate questions for short Wikipedia-like articles")
|
11 |
+
|
12 |
+
st_model_load = st.text('Loading question generator model...')
|
13 |
+
|
14 |
+
@st.cache(allow_output_mutation=True)
|
15 |
+
def load_model():
|
16 |
+
print("Loading model...")
|
17 |
+
tokenizer = T5TokenizerFast.from_pretrained("t5-base")
|
18 |
+
model = T5ForConditionalGeneration.from_pretrained(model_name)
|
19 |
+
nltk.download('punkt')
|
20 |
+
print("Model loaded!")
|
21 |
+
return tokenizer, model
|
22 |
+
|
23 |
+
tokenizer, model = load_model()
|
24 |
+
st.success('Model loaded!')
|
25 |
+
st_model_load.text("")
|
26 |
+
|
27 |
+
if 'text' not in st.session_state:
|
28 |
+
st.session_state.text = ""
|
29 |
+
st_text_area = st.text_area('Text to generate the questions for', value=st.session_state.text, height=500)
|
30 |
+
|
31 |
+
def generate_questions():
|
32 |
+
st.session_state.text = st_text_area
|
33 |
+
|
34 |
+
generator_args = {
|
35 |
+
"max_length": 256,
|
36 |
+
"num_beams": 4,
|
37 |
+
"length_penalty": 1.5,
|
38 |
+
"no_repeat_ngram_size": 3,
|
39 |
+
"early_stopping": True,
|
40 |
+
}
|
41 |
+
input_string = "generate questions: " + st_text_area + " </s>"
|
42 |
+
input_ids = tokenizer.encode(input_string, return_tensors="pt")
|
43 |
+
res = model.generate(input_ids, **generator_args)
|
44 |
+
output = tokenizer.batch_decode(res, skip_special_tokens=True)
|
45 |
+
output = [question.strip() + "?" for question in output[0].split("?") if question != ""]
|
46 |
+
|
47 |
+
st.session_state.questions = output
|
48 |
+
|
49 |
+
# generate title button
|
50 |
+
st_generate_button = st.button('Generate questions', on_click=generate_questions)
|
51 |
+
|
52 |
+
# title generation labels
|
53 |
+
if 'questions' not in st.session_state:
|
54 |
+
st.session_state.questions = []
|
55 |
+
|
56 |
+
if len(st.session_state.questions) > 0:
|
57 |
+
with st.container():
|
58 |
+
st.subheader("Generated questions")
|
59 |
+
for title in st.session_state.questions:
|
60 |
+
st.markdown("__" + title + "__")
|