Update app.py

app.py

@@ -1,33 +1,24 @@
-import os
-import sys
-from subprocess import check_call
-
-# Ensure transformers is installed
-try:
-    from transformers import DetrImageProcessor, DetrForObjectDetection, pipeline
-except ImportError:
-    check_call([sys.executable, "-m", "pip", "install", "transformers==4.33.2"])
-    from transformers import DetrImageProcessor, DetrForObjectDetection, pipeline
-
+from transformers import DetrImageProcessor, DetrForObjectDetection, pipeline
 from PIL import Image
 import gradio as gr
 
 # Load pre-trained models
 detection_model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")
 detection_processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50")
-description_generator = pipeline("text-generation", model="gpt-2")
 
-# Function to recognize and describe a product
+# Function to process image and generate description
 def recognize_and_describe(image):
+    # Process the image with DETR
     inputs = detection_processor(images=image, return_tensors="pt")
     outputs = detection_model(**inputs)
-    logits = outputs.logits.argmax(-1).item()
-    product_label = f"Product Class: {logits}"
-
-    # Generate description
-    prompt = f"Describe the product: {product_label}"
-    description = description_generator(prompt, max_length=50, num_return_sequences=1)
-    return product_label, description[0]["generated_text"]
+    
+    # Get detected classes
+    logits = outputs.logits.argmax(-1).tolist()[0]
+    product_label = f"Detected Product Class: {logits}"
+    
+    # Generate a description using a dummy model or hardcoded description
+    description = f"This is a product in class {logits}. Further information can be retrieved."
+    return product_label, description
 
 # Gradio Interface
 interface = gr.Interface(