Update app.py

app.py

@@ -1,36 +1,36 @@
 import gradio as gr
-from PIL import Image
 import requests
-import io
-
-# Function to perform OCR using Qwen model
-def perform_ocr(image):
-    # Convert the image to bytes
-    buffered = io.BytesIO()
-    image.save(buffered, format="JPEG")
-    image_bytes = buffered.getvalue()
-    
-    # API URL for Qwen OCR model
-    api_url = "https://api-inference.huggingface.co/models/Qwen/Qwen-VL"
-    headers = {"Authorization": f"Bearer {HF}"}
+import os
+
+# Function to query the Qwen model for OCR
+def query_ocr(image):
+    # Get the Hugging Face access token from secrets
+    hf_token = os.getenv("secret")
+    headers = {"Authorization": f"Bearer {hf_token}"}
+
+    # API endpoint for the model
+    API_URL = "https://api-inference.huggingface.co/models/Qwen/Qwen-VL"
 
-    # Make a request to the model
-    response = requests.post(api_url, headers=headers, files={"file": image_bytes})
+    # Prepare the request
+    files = {"file": image}
+    response = requests.post(API_URL, headers=headers, files=files)
+
+    # Raise an error if the request fails
+    response.raise_for_status()
     
-    # Check if the request was successful
-    if response.status_code == 200:
-        result = response.json()
-        return result['text']  # Adjust according to the API response format
-    else:
-        return "Error: " + response.text
+    return response.json()  # Assuming the model returns a JSON with the extracted text
 
 # Gradio interface
 with gr.Blocks() as demo:
-    gr.Markdown("## OCR with Qwen Model")
-    image_input = gr.Image(type="pil", label="Upload Image")
-    ocr_button = gr.Button("Perform OCR")
-    text_output = gr.Textbox(label="Extracted Text", lines=10)
+    gr.Markdown("# OCR Application using Qwen Model")
+    gr.Markdown("Upload an image to extract text using OCR.")
+
+    image_input = gr.Image(type="file", label="Upload Image")
+    output_text = gr.Textbox(label="Extracted Text", lines=10)
 
-    ocr_button.click(perform_ocr, inputs=image_input, outputs=text_output)
+    submit_btn = gr.Button("Extract Text")
+    
+    submit_btn.click(fn=query_ocr, inputs=image_input, outputs=output_text)
 
+# Launch the app
 demo.launch()