Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,3 +1,4 @@
|
|
|
|
|
| 1 |
import streamlit as st
|
| 2 |
from transformers import pipeline, AutoTokenizer, AutoModelForQuestionAnswering
|
| 3 |
|
|
@@ -12,10 +13,20 @@ st.sidebar.header('What will you like to do?')
|
|
| 12 |
option = st.sidebar.radio('', ['Extractive question answering', 'Text summarization', 'Text generation', 'Sentiment analysis'])
|
| 13 |
|
| 14 |
@st.cache(show_spinner=False, allow_output_mutation=True)
|
| 15 |
-
def
|
| 16 |
tokenizer = AutoTokenizer.from_pretrained("deepset/roberta-base-squad2")
|
| 17 |
-
|
| 18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
@st.cache(show_spinner=False, allow_output_mutation=True)
|
| 21 |
def summarization_model():
|
|
@@ -41,10 +52,9 @@ if option == 'Extractive question answering':
|
|
| 41 |
question = st.text_input(label='Enter your question')
|
| 42 |
button = st.button('Get answer')
|
| 43 |
if button:
|
| 44 |
-
question_answerer = question_model()
|
| 45 |
with st.spinner(text="Getting answer..."):
|
| 46 |
-
answer = question_answerer(context
|
| 47 |
-
st.write(answer
|
| 48 |
elif source == "I want to upload a file":
|
| 49 |
uploaded_file = st.file_uploader("Choose a .txt file to upload", type=["txt"])
|
| 50 |
question = st.text_input(label='Enter your question')
|
|
|
|
| 1 |
+
import torch
|
| 2 |
import streamlit as st
|
| 3 |
from transformers import pipeline, AutoTokenizer, AutoModelForQuestionAnswering
|
| 4 |
|
|
|
|
| 13 |
option = st.sidebar.radio('', ['Extractive question answering', 'Text summarization', 'Text generation', 'Sentiment analysis'])
|
| 14 |
|
| 15 |
@st.cache(show_spinner=False, allow_output_mutation=True)
|
| 16 |
+
def question_answerer(context, question):
|
| 17 |
tokenizer = AutoTokenizer.from_pretrained("deepset/roberta-base-squad2")
|
| 18 |
+
model = AutoModelForQuestionAnswering.from_pretrained("deepset/roberta-base-squad2")
|
| 19 |
+
inputs = tokenizer(question, context, add_special_tokens=True, return_tensors="pt")
|
| 20 |
+
input_ids = inputs["input_ids"].tolist()[0]
|
| 21 |
+
outputs = model(**inputs)
|
| 22 |
+
answer_start_scores = outputs.start_logits
|
| 23 |
+
answer_end_scores = outputs.end_logits
|
| 24 |
+
# Get the most likely beginning of answer with the argmax of the score
|
| 25 |
+
answer_start = torch.argmax(answer_start_scores)
|
| 26 |
+
# Get the most likely end of answer with the argmax of the score
|
| 27 |
+
answer_end = torch.argmax(answer_end_scores) + 1
|
| 28 |
+
answer = tokenizer.convert_tokens_to_string(tokenizer.convert_ids_to_tokens(input_ids[answer_start:answer_end]))
|
| 29 |
+
return answer
|
| 30 |
|
| 31 |
@st.cache(show_spinner=False, allow_output_mutation=True)
|
| 32 |
def summarization_model():
|
|
|
|
| 52 |
question = st.text_input(label='Enter your question')
|
| 53 |
button = st.button('Get answer')
|
| 54 |
if button:
|
|
|
|
| 55 |
with st.spinner(text="Getting answer..."):
|
| 56 |
+
answer = question_answerer(context, question)
|
| 57 |
+
st.write(answer)
|
| 58 |
elif source == "I want to upload a file":
|
| 59 |
uploaded_file = st.file_uploader("Choose a .txt file to upload", type=["txt"])
|
| 60 |
question = st.text_input(label='Enter your question')
|