Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
|
@@ -2,8 +2,7 @@ import streamlit as st
|
|
| 2 |
from streamlit.components.v1 import html
|
| 3 |
import os
|
| 4 |
import PyPDF2
|
| 5 |
-
|
| 6 |
-
st.write(os.environ("TEST"))
|
| 7 |
|
| 8 |
def get_pdf_text(pdf_path):
|
| 9 |
# creating a pdf file object
|
|
@@ -24,6 +23,22 @@ def get_pdf_text(pdf_path):
|
|
| 24 |
|
| 25 |
return pdf_text
|
| 26 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
tab_general_topics, tab_your_paper = st.tabs(["Research topics", "Summarize your paper(s)"])
|
| 28 |
|
| 29 |
with tab_general_topics:
|
|
@@ -46,7 +61,7 @@ with tab_your_paper:
|
|
| 46 |
for pdf in pdf_files:
|
| 47 |
# Saving the files
|
| 48 |
pdf_data = pdf.getvalue()
|
| 49 |
-
pdf_path = os.path.join(pdf.name)
|
| 50 |
with open(pdf_path, "wb") as f:
|
| 51 |
f.write(pdf_data)
|
| 52 |
recently_added.append(pdf_path)
|
|
@@ -55,6 +70,7 @@ with tab_your_paper:
|
|
| 55 |
for recent_pdf in recently_added:
|
| 56 |
# Reading the pdf files
|
| 57 |
pdf_content = get_pdf_text(recent_pdf)
|
|
|
|
| 58 |
pdfs_content_list.append(pdf_content)
|
| 59 |
|
| 60 |
# Delete the files
|
|
@@ -62,4 +78,33 @@ with tab_your_paper:
|
|
| 62 |
|
| 63 |
all_text_together = " ".join(pdfs_content_list)
|
| 64 |
|
| 65 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
from streamlit.components.v1 import html
|
| 3 |
import os
|
| 4 |
import PyPDF2
|
| 5 |
+
import requests
|
|
|
|
| 6 |
|
| 7 |
def get_pdf_text(pdf_path):
|
| 8 |
# creating a pdf file object
|
|
|
|
| 23 |
|
| 24 |
return pdf_text
|
| 25 |
|
| 26 |
+
headers = {"Authorization": "Bearer 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"}
|
| 27 |
+
|
| 28 |
+
def create_tags(payload):
|
| 29 |
+
API_URL_TAGS = "https://api-inference.huggingface.co/models/fabiochiu/t5-base-tag-generation"
|
| 30 |
+
|
| 31 |
+
response = requests.post(API_URL_TAGS, headers=headers, json=payload)
|
| 32 |
+
return response.json()
|
| 33 |
+
|
| 34 |
+
def summarize_text(payload):
|
| 35 |
+
API_URL = "https://api-inference.huggingface.co/models/facebook/bart-large-cnn"
|
| 36 |
+
response = requests.post(API_URL, headers=headers, json=payload)
|
| 37 |
+
return response.json()
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
# Start of the app code
|
| 41 |
+
|
| 42 |
tab_general_topics, tab_your_paper = st.tabs(["Research topics", "Summarize your paper(s)"])
|
| 43 |
|
| 44 |
with tab_general_topics:
|
|
|
|
| 61 |
for pdf in pdf_files:
|
| 62 |
# Saving the files
|
| 63 |
pdf_data = pdf.getvalue()
|
| 64 |
+
pdf_path = os.path.join("pdfs", pdf.name)
|
| 65 |
with open(pdf_path, "wb") as f:
|
| 66 |
f.write(pdf_data)
|
| 67 |
recently_added.append(pdf_path)
|
|
|
|
| 70 |
for recent_pdf in recently_added:
|
| 71 |
# Reading the pdf files
|
| 72 |
pdf_content = get_pdf_text(recent_pdf)
|
| 73 |
+
print("**", pdf_content)
|
| 74 |
pdfs_content_list.append(pdf_content)
|
| 75 |
|
| 76 |
# Delete the files
|
|
|
|
| 78 |
|
| 79 |
all_text_together = " ".join(pdfs_content_list)
|
| 80 |
|
| 81 |
+
try:
|
| 82 |
+
tags = create_tags({
|
| 83 |
+
"inputs": all_text_together,
|
| 84 |
+
})[0]["generated_text"]
|
| 85 |
+
tags_available = True
|
| 86 |
+
except:
|
| 87 |
+
tags_available = False
|
| 88 |
+
|
| 89 |
+
summary = summarize_text({
|
| 90 |
+
"inputs": all_text_together
|
| 91 |
+
})[0]["summary_text"]
|
| 92 |
+
|
| 93 |
+
col1, col2 = st.columns(2)
|
| 94 |
+
with col1:
|
| 95 |
+
if len(recently_added) > 1:
|
| 96 |
+
st.markdown("#### Summary of your paper(s):")
|
| 97 |
+
else:
|
| 98 |
+
st.markdown("#### Summary of your paper:")
|
| 99 |
+
st.write(summary)
|
| 100 |
+
|
| 101 |
+
if tags_available == True:
|
| 102 |
+
with col2:
|
| 103 |
+
if len(recently_added) > 1:
|
| 104 |
+
st.markdown("#### Identified topics of your paper(s):")
|
| 105 |
+
else:
|
| 106 |
+
st.markdown("#### Identified topics of your paper:")
|
| 107 |
+
st.write(tags)
|
| 108 |
+
|
| 109 |
+
with st.expander("See your total text"):
|
| 110 |
+
st.write(all_text_together)
|