jayparmar0109 commited on
Commit
11e1c78
·
verified ·
1 Parent(s): 2aff8a5

Create app.py

Browse files
Files changed (1) hide show
  1. app.py +65 -0
app.py ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import streamlit as st
2
+ import os
3
+ import pathlib
4
+ import textwrap
5
+ from PIL import Image
6
+
7
+
8
+ import google.generativeai as genai
9
+
10
+ genai.configure(api_key='AIzaSyCeNgXfZx0kJ736XFVtxXxev_RdscB0i5s')
11
+
12
+ ## Function to load OpenAI model and get respones
13
+
14
+ def get_gemini_response(input,image,prompt):
15
+ model = genai.GenerativeModel('gemini-pro-vision')
16
+ response = model.generate_content([input,image[0],prompt])
17
+ return response.text
18
+
19
+
20
+ def input_image_setup(uploaded_file):
21
+ # Check if a file has been uploaded
22
+ if uploaded_file is not None:
23
+ # Read the file into bytes
24
+ bytes_data = uploaded_file.getvalue()
25
+
26
+ image_parts = [
27
+ {
28
+ "mime_type": uploaded_file.type, # Get the mime type of the uploaded file
29
+ "data": bytes_data
30
+ }
31
+ ]
32
+ return image_parts
33
+ else:
34
+ raise FileNotFoundError("No file uploaded")
35
+
36
+
37
+ ##initialize our streamlit app
38
+
39
+ st.set_page_config(page_title="Gemini Image Demo")
40
+
41
+ st.header("Generative AI : Invoice Reader")
42
+ input=st.text_input("Input Prompt: ",key="input")
43
+ uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])
44
+ image=""
45
+ if uploaded_file is not None:
46
+ image = Image.open(uploaded_file)
47
+ st.image(image, caption="Uploaded Image.", use_column_width=True)
48
+
49
+
50
+ submit=st.button("Tell me about the image")
51
+
52
+ input_prompt = """
53
+ You are an expert in understanding business cards.
54
+ You will receive input images as business card & you will have to answer questions based on the input image.
55
+ You have to extract information from business card images and give correct tag to the output text
56
+ like person name, company name, occupation, address, phone/telephone number, email, website, etc.
57
+ """
58
+
59
+ ## If ask button is clicked
60
+
61
+ if submit:
62
+ image_data = input_image_setup(uploaded_file)
63
+ response=get_gemini_response(input_prompt,image_data,input)
64
+ st.subheader("The Response is")
65
+ st.write(response)