Spaces:
Build error
Build error
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,5 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
def chunk_text(text, max_tokens=5000):
|
| 2 |
-
"""Split text into smaller chunks."""
|
| 3 |
words = text.split()
|
| 4 |
chunks = []
|
| 5 |
current_chunk = []
|
|
@@ -17,23 +39,51 @@ def chunk_text(text, max_tokens=5000):
|
|
| 17 |
chunks.append(" ".join(current_chunk))
|
| 18 |
return chunks
|
| 19 |
|
|
|
|
| 20 |
def generate_questions_from_text(text):
|
| 21 |
-
"""Generate questions by processing text in chunks."""
|
| 22 |
chunks = chunk_text(text, max_tokens=5000)
|
| 23 |
all_questions = []
|
| 24 |
|
| 25 |
-
for chunk in chunks:
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
|
| 39 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
import streamlit as st
|
| 3 |
+
from groq import Groq
|
| 4 |
+
import PyPDF2
|
| 5 |
+
|
| 6 |
+
# Initialize Groq client
|
| 7 |
+
client = Groq(
|
| 8 |
+
api_key=os.environ.get("GROQ_API_KEY"),
|
| 9 |
+
)
|
| 10 |
+
|
| 11 |
+
# Function to extract text from uploaded PDF
|
| 12 |
+
def extract_text_from_pdf(pdf_file):
|
| 13 |
+
try:
|
| 14 |
+
pdf_reader = PyPDF2.PdfReader(pdf_file)
|
| 15 |
+
text = ""
|
| 16 |
+
for page in pdf_reader.pages:
|
| 17 |
+
text += page.extract_text()
|
| 18 |
+
return text
|
| 19 |
+
except Exception as e:
|
| 20 |
+
st.error(f"Error reading PDF: {e}")
|
| 21 |
+
return ""
|
| 22 |
+
|
| 23 |
+
# Function to split text into smaller chunks
|
| 24 |
def chunk_text(text, max_tokens=5000):
|
|
|
|
| 25 |
words = text.split()
|
| 26 |
chunks = []
|
| 27 |
current_chunk = []
|
|
|
|
| 39 |
chunks.append(" ".join(current_chunk))
|
| 40 |
return chunks
|
| 41 |
|
| 42 |
+
# Function to generate questions from text using Groq API
|
| 43 |
def generate_questions_from_text(text):
|
|
|
|
| 44 |
chunks = chunk_text(text, max_tokens=5000)
|
| 45 |
all_questions = []
|
| 46 |
|
| 47 |
+
for i, chunk in enumerate(chunks):
|
| 48 |
+
try:
|
| 49 |
+
chat_completion = client.chat.completions.create(
|
| 50 |
+
messages=[
|
| 51 |
+
{
|
| 52 |
+
"role": "user",
|
| 53 |
+
"content": f"Generate a list of questions based on the following content:\n{chunk}",
|
| 54 |
+
}
|
| 55 |
+
],
|
| 56 |
+
model="llama-3.3-70b-versatile",
|
| 57 |
+
stream=False,
|
| 58 |
+
)
|
| 59 |
+
questions = chat_completion.choices[0].message.content
|
| 60 |
+
all_questions.append(f"Chunk {i+1} Questions:\n{questions}")
|
| 61 |
+
except Exception as e:
|
| 62 |
+
st.error(f"Error processing chunk {i+1}: {e}")
|
| 63 |
+
|
| 64 |
+
return "\n\n".join(all_questions)
|
| 65 |
+
|
| 66 |
+
# Streamlit app UI and logic
|
| 67 |
+
st.title("Generate Questions from Your PDF")
|
| 68 |
+
|
| 69 |
+
uploaded_file = st.file_uploader("Upload a PDF file", type="pdf")
|
| 70 |
+
|
| 71 |
+
if uploaded_file is not None:
|
| 72 |
+
st.write("### Uploaded PDF:")
|
| 73 |
+
text = extract_text_from_pdf(uploaded_file)
|
| 74 |
|
| 75 |
+
if not text.strip():
|
| 76 |
+
st.error("No text could be extracted from the uploaded PDF.")
|
| 77 |
+
else:
|
| 78 |
+
st.text_area("Extracted Text", text, height=300)
|
| 79 |
+
|
| 80 |
+
if st.button("Generate Questions"):
|
| 81 |
+
with st.spinner("Generating questions..."):
|
| 82 |
+
questions = generate_questions_from_text(text)
|
| 83 |
+
if questions.strip():
|
| 84 |
+
st.write("### Generated Questions:")
|
| 85 |
+
st.write(questions)
|
| 86 |
+
else:
|
| 87 |
+
st.error("No questions could be generated from the content.")
|
| 88 |
+
else:
|
| 89 |
+
st.info("Please upload a PDF to start.")
|