Update app.py

app.py

@@ -14,8 +14,8 @@ model = AutoModelForCausalLM.from_pretrained('HuggingFaceM4/Florence-2-DocVQA',
 processor = AutoProcessor.from_pretrained('HuggingFaceM4/Florence-2-DocVQA', trust_remote_code=True)
 
 
-TITLE = "# [Descripta Demo](https://huggingface.co/HuggingFaceM4/Florence-2-DocVQA)"
-DESCRIPTION = "The demo for Florence-2 fine-tuned on DocVQA dataset. You can find the notebook [here](https://colab.research.google.com/drive/1hKDrJ5AH_o7I95PtZ9__VlCTNAo1Gjpf?usp=sharing). Read more about Florence-2 fine-tuning [here](finetune-florence2)."
+TITLE = "#Descripta"
+DESCRIPTION = "Generate high-quality product descriptions from images and metadata"
 
 
 colormap = ['blue','orange','green','purple','brown','pink','gray','olive','cyan','red',
@@ -74,9 +74,9 @@ with gr.Blocks(css=css) as demo:
         gr.Examples(
             examples=[
                 ["hunt.jpg", 'What is this image?'],
-                ["idefics2_architecture.png", 'How many tokens per image does it use?'],
-                ["idefics2_architecture.png", "What type of encoder does the model use?"],
-                ["image.jpg", "What's the share of Industry Switchers Gained?"]
+                # ["idefics2_architecture.png", 'How many tokens per image does it use?'],
+                # ["idefics2_architecture.png", "What type of encoder does the model use?"],
+                # ["image.jpg", "What's the share of Industry Switchers Gained?"]
             ],
             inputs=[input_img, text_input],
             outputs=[output_text],