sync wtih remote
Browse files- app.py +69 -0
- requirements.txt +2 -0
app.py
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
import tempfile
|
3 |
+
from PIL import Image
|
4 |
+
import google.generativeai as genai
|
5 |
+
import os
|
6 |
+
|
7 |
+
MODEL_ID = "gemini-2.0-flash-exp"
|
8 |
+
api_key = os.getenv("GEMINI_API_KEY")
|
9 |
+
model_id = MODEL_ID
|
10 |
+
genai.configure(api_key=api_key)
|
11 |
+
|
12 |
+
if "model" not in st.session_state:
|
13 |
+
st.session_state.model = genai.GenerativeModel(MODEL_ID)
|
14 |
+
|
15 |
+
model = st.session_state.model
|
16 |
+
chat = model.start_chat()
|
17 |
+
mime_type = "image/jpeg" # Use a consistent MIME type for images
|
18 |
+
|
19 |
+
|
20 |
+
|
21 |
+
# Dummy AI model function
|
22 |
+
def ai_model(image_path):
|
23 |
+
"""
|
24 |
+
This is a dummy AI model function that will be replaced with
|
25 |
+
the actual multimodal AI model.
|
26 |
+
"""
|
27 |
+
return "The AI model response based on the image."
|
28 |
+
|
29 |
+
def main():
|
30 |
+
st.title("Image Processing with AI")
|
31 |
+
img_file_buffer = None
|
32 |
+
image_path = ""
|
33 |
+
|
34 |
+
# Use the camera to capture an image
|
35 |
+
img_file_buffer = st.camera_input("Take a picture")
|
36 |
+
|
37 |
+
if img_file_buffer is not None:
|
38 |
+
# Save the image to a temporary file
|
39 |
+
with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as temp_file:
|
40 |
+
temp_file.write(img_file_buffer.read())
|
41 |
+
image_path = temp_file.name
|
42 |
+
st.write("Image saved to:", image_path)
|
43 |
+
#st.image(Image.open(image_path), caption="Captured Image", use_container_width=True)
|
44 |
+
|
45 |
+
user_input = st.text_area("Enter your prompt:", height=200)
|
46 |
+
prompt =f"Refer to the image. Perform the task or answer the question {user_input}"
|
47 |
+
|
48 |
+
if st.button("Process the image"):
|
49 |
+
if prompt:
|
50 |
+
with st.spinner("Processing..."):
|
51 |
+
# Upload the file with the correct MIME type
|
52 |
+
file_data = genai.upload_file(image_path, mime_type=mime_type)
|
53 |
+
|
54 |
+
# Send file and prompt to Gemini API
|
55 |
+
response = chat.send_message(
|
56 |
+
[
|
57 |
+
prompt,
|
58 |
+
file_data
|
59 |
+
]
|
60 |
+
)
|
61 |
+
|
62 |
+
# Display Gemini response
|
63 |
+
st.markdown(response.text)
|
64 |
+
else:
|
65 |
+
st.warning("Please provide a valid prompt.")
|
66 |
+
|
67 |
+
|
68 |
+
if __name__ == "__main__":
|
69 |
+
main()
|
requirements.txt
ADDED
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
1 |
+
streamlit
|
2 |
+
google-generativeai
|