Update app.py

app.py

@@ -1,32 +1,27 @@
-import streamlit as st
-from PIL import Image
-from PIL import Image, ImageDraw
-from image_whisper_helper import summarize_predictions_natural_language, render_results_in_image
+import gradio as gr
 from transformers import pipeline
-from tokenizers import Tokenizer, Encoding
-from tokenizers import decoders
-from tokenizers import models
-from tokenizers import normalizers
-from tokenizers import pre_tokenizers
-from tokenizers import processors
-import io
-import matplotlib.pyplot as plt
-import requests
-import inflect
-from PIL import Image
-from predictions import get_predictions  # Replace 'your_module' with the name of the module where your function is defined
 
-def main():
-    st.title("Object Detection App")
+# Load the image-to-text pipeline
+image_to_text_pipelines = {
+    "Salesforce/blip-image-captioning-base": pipeline("image-to-text", model="Salesforce/blip-image-captioning-base"),
+    # Add more models if needed
+}
 
-    uploaded_image = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
+def generate_caption(input_image, model_name="Salesforce/blip-image-captioning-base"):
+    # Generate caption for the input image using the selected model
+    image_to_text_pipeline = image_to_text_pipelines[model_name]
+    caption = image_to_text_pipeline(input_image)[0]['generated_text']
+    return caption
 
-    if uploaded_image is not None:
-        processed_image, text, audio = get_predictions(uploaded_image)
+# Interface for launching the model
+interface = gr.Interface(
+    fn=generate_caption,
+    inputs=gr.Image(type='pil', label="Input Image"),
+    outputs="text",
+    title="Image Captioning Model",
+    description="This model generates captions for images.",
+    theme="default",
+)
 
-        st.image(processed_image, caption='Processed Image', use_column_width=True)
-        st.write(f"Predictions: {text}")
-        st.audio(audio, format='audio/wav')
-
-if __name__ == '__main__':
-    main()
+# Launch the interface
+interface.launch()