Spaces:
Sleeping
Sleeping
File size: 4,052 Bytes
f76d77f ea6a4e1 f76d77f 047420b f76d77f 047420b f76d77f 27255d4 047420b f76d77f a2d596c 4996d49 a2d596c 4996d49 a2d596c 4996d49 a2d596c 8e253a5 4996d49 8e253a5 4996d49 a2d596c 4996d49 a2d596c 4996d49 a2d596c 4996d49 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 |
import streamlit as st
import base64
import openai
# Function to encode the image to base64
def encode_image(image_file):
return base64.b64encode(image_file.getvalue()).decode("utf-8")
# Streamlit page setup
st.set_page_config(page_title="MTSS Image Accessibility Alt Text Generator", layout="centered", initial_sidebar_state="collapsed")
st.title("MTSS Snapshot: Accessibility Image Textifier: `Alt Text`")
# Retrieve the OpenAI API key from Streamlit secrets and set it
openai.api_key = st.secrets["openai_api_key"]
# Set the OpenAI API key for the client
openai.api_key = api_key
# File uploader for images
uploaded_file = st.file_uploader("Upload an image", type=["jpg", "png", "jpeg"])
if uploaded_file:
with st.expander("Image", expanded=True):
st.image(uploaded_file, caption=uploaded_file.name, use_column_width=True)
# Toggle for additional details input
show_details = st.checkbox("Add details about the image")
if show_details:
# Text input for additional details about the image, shown only if toggle is True
additional_details = st.text_area(
"Add any additional details or context about the image here:",
disabled=not show_details
)
# Button to trigger the analysis
analyze_button = st.button("Analyse the MTSS Image")
# Check if an image has been uploaded, if the API key is available, and if the button has been pressed
if uploaded_file is not None and api_key and analyze_button:
with st.spinner("Analysing the image ..."):
# Encode the image
base64_image = encode_image(uploaded_file)
# Optimized prompt for additional clarity and detail
prompt_text = (
"You are a highly knowledgeable accessibility expert. "
"Your task is to examine the following image in detail. "
"Provide a comprehensive, factual, and accurate explanation of what the image depicts. "
"Highlight key elements and their significance, and present your analysis in clear, well-structured format. "
"Create a detailed image caption explaining in 150 words or less. "
)
if show_details and additional_details:
prompt_text += f"\n\nAdditional Context Provided by the User:\n{additional_details}"
# Create the payload for the completion request
messages = [
{
"role": "user",
"content": [
{"type": "text", "text": prompt_text},
{
"type": "image_url",
"image_url": f"data:image/jpeg;base64,{base64_image}",
},
],
}
]
# Make the request to the OpenAI API
try:
# Without Stream
# response = openai.chat.completions.create(
# model="gpt-4-vision-preview", messages=messages, max_tokens=500, stream=False
# )
# Stream the response
full_response = ""
message_placeholder = st.empty()
for completion in openai.chat.completions.create(
model="gpt-4-vision-preview", messages=messages,
max_tokens=150, stream=True
):
# Check if there is content to display
if completion.choices[0].delta.content is not None:
full_response += completion.choices[0].delta.content
message_placeholder.markdown(full_response + "▌")
# Final update to placeholder after the stream ends
message_placeholder.markdown(full_response)
# Display the response in the app
# st.write(response.choices[0].message.content)
except Exception as e:
st.error(f"An error occurred: {e}")
else:
# Warnings for user action required
if not uploaded_file and analyze_button:
st.warning("Please upload an image.")
if not api_key:
st.warning("Please enter your OpenAI API key.")
|