Spaces:
Sleeping
Sleeping
add pdf ext code
Browse files
app.py
CHANGED
|
@@ -1,5 +1,7 @@
|
|
| 1 |
import gradio as gr
|
| 2 |
from huggingface_hub import InferenceClient
|
|
|
|
|
|
|
| 3 |
|
| 4 |
"""
|
| 5 |
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
|
|
@@ -40,6 +42,20 @@ def respond(
|
|
| 40 |
yield response
|
| 41 |
|
| 42 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
"""
|
| 44 |
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
|
| 45 |
"""
|
|
@@ -59,6 +75,18 @@ demo = gr.ChatInterface(
|
|
| 59 |
],
|
| 60 |
)
|
| 61 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
|
| 63 |
if __name__ == "__main__":
|
| 64 |
demo.launch()
|
|
|
|
| 1 |
import gradio as gr
|
| 2 |
from huggingface_hub import InferenceClient
|
| 3 |
+
import PyPDF2
|
| 4 |
+
import io
|
| 5 |
|
| 6 |
"""
|
| 7 |
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
|
|
|
|
| 42 |
yield response
|
| 43 |
|
| 44 |
|
| 45 |
+
def extract_text_from_pdf(pdf_file):
|
| 46 |
+
if pdf_file is None:
|
| 47 |
+
return "No file uploaded."
|
| 48 |
+
|
| 49 |
+
try:
|
| 50 |
+
pdf_reader = PyPDF2.PdfReader(io.BytesIO(pdf_file))
|
| 51 |
+
text = ""
|
| 52 |
+
for page in pdf_reader.pages:
|
| 53 |
+
text += page.extract_text() + "\n\n"
|
| 54 |
+
return text.strip()
|
| 55 |
+
except Exception as e:
|
| 56 |
+
return f"An error occurred: {str(e)}"
|
| 57 |
+
|
| 58 |
+
|
| 59 |
"""
|
| 60 |
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
|
| 61 |
"""
|
|
|
|
| 75 |
],
|
| 76 |
)
|
| 77 |
|
| 78 |
+
pdf_interface = gr.Interface(
|
| 79 |
+
fn=extract_text_from_pdf,
|
| 80 |
+
inputs=gr.File(label="Upload PDF", type="binary"),
|
| 81 |
+
outputs="text",
|
| 82 |
+
title="PDF Text Extractor",
|
| 83 |
+
description="Upload a PDF file to extract its text content."
|
| 84 |
+
)
|
| 85 |
+
|
| 86 |
+
demo = gr.TabbedInterface(
|
| 87 |
+
[demo, pdf_interface],
|
| 88 |
+
["Chat", "PDF Extractor"]
|
| 89 |
+
)
|
| 90 |
|
| 91 |
if __name__ == "__main__":
|
| 92 |
demo.launch()
|