Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
from transformers import pipeline
|
| 3 |
+
import PyPDF2
|
| 4 |
+
|
| 5 |
+
# Load the explainer model
|
| 6 |
+
explainer = pipeline("text2text-generation", model="google/flan-t5-base")
|
| 7 |
+
|
| 8 |
+
def extract_text_from_pdf(pdf_file):
|
| 9 |
+
text = ""
|
| 10 |
+
try:
|
| 11 |
+
with open(pdf_file.name, 'rb') as pdfFileObj:
|
| 12 |
+
pdfReader = PyPDF2.PdfReader(pdfFileObj)
|
| 13 |
+
for pageNum in range(len(pdfReader.pages)):
|
| 14 |
+
pageObj = pdfReader.pages[pageNum]
|
| 15 |
+
text += pageObj.extract_text()
|
| 16 |
+
except Exception as e:
|
| 17 |
+
return f"Error reading PDF: {e}"
|
| 18 |
+
return text
|
| 19 |
+
|
| 20 |
+
def process_and_explain(pdf_file, concept_to_explain):
|
| 21 |
+
if pdf_file is not None:
|
| 22 |
+
extracted_text = extract_text_from_pdf(pdf_file)
|
| 23 |
+
if "Error reading PDF" in extracted_text or not extracted_text.strip():
|
| 24 |
+
return "Could not extract text from the PDF. Please try another file."
|
| 25 |
+
|
| 26 |
+
prompt = f"Explain the concept of '{concept_to_explain}' based on this text: '{extracted_text[:1500]}...' in simple terms." # Limiting text for the model
|
| 27 |
+
explanation_result = explainer(prompt, max_length=300, do_sample=False)
|
| 28 |
+
explanation = explanation_result[0]["generated_text"]
|
| 29 |
+
return f"**Extracted Text Snippet:**\n{extracted_text[:500]}...\n\n**Explanation of '{concept_to_explain}':**\n{explanation}"
|
| 30 |
+
else:
|
| 31 |
+
return "Please upload a PDF file."
|
| 32 |
+
|
| 33 |
+
with gr.Blocks() as demo:
|
| 34 |
+
gr.Markdown("## Explain Concepts from Uploaded PDFs")
|
| 35 |
+
gr.Markdown("Upload a PDF, and then specify a concept you'd like explained based on its content.")
|
| 36 |
+
|
| 37 |
+
pdf_input = gr.File(label="Upload PDF", file_types=[".pdf"])
|
| 38 |
+
concept_input = gr.Textbox(label="Concept to Explain", placeholder="E.g., Climate Change")
|
| 39 |
+
explain_button = gr.Button("Explain Concept")
|
| 40 |
+
output_text = gr.Markdown(label="Explanation")
|
| 41 |
+
|
| 42 |
+
explain_button.click(
|
| 43 |
+
fn=process_and_explain,
|
| 44 |
+
inputs=[pdf_input, concept_input],
|
| 45 |
+
outputs=output_text
|
| 46 |
+
)
|
| 47 |
+
|
| 48 |
+
demo.launch()
|