Update app.py

app.py

@@ -98,9 +98,10 @@ prompt_suffix = "\n"
 
 # Vision model tab function
 @spaces.GPU()
-def stream_vision(image, text_input=None, model_id="justinj92/phi-35-vision-burberry"):
+def stream_vision(image, model_id="justinj92/phi-35-vision-burberry"):
     model = models[model_id]
     processor = processors[model_id]
+    text_input="What is shown in this image?"
 
     # Prepare the image list and corresponding tags
     images = [Image.fromarray(image).convert("RGB")]
@@ -128,7 +129,7 @@ def stream_vision(image, text_input=None, model_id="justinj92/phi-35-vision-burb
 
     # Generation parameters
     generation_args = {
-        "max_new_tokens": 1000,
+        "max_new_tokens": 2000,
         "temperature": 0.0,
         "do_sample": False,
     }
@@ -165,18 +166,12 @@ h3 {
 }
 """
 
-PLACEHOLDER = """
-<center>
-<p>Hi! I'm your assistant. Feel free to ask your questions</p>
-</center>
-"""
-
-TITLE = "<h1><center>Phi-3.5 Fine-Tuned Vision Bot using Burberry Dataset </center></h1>"
+TITLE = "<h1><center>Burberry Product Categorizer</center></h1>"
 
 EXPLANATION = """
 <div style="text-align: center; margin-top: 20px;">
-    <p>This app supports microsoft/Phi-3.5-vision-instruct model for multimodal model.</p>
-    <p>This Fine-Tuned version is built using open Burberry Product dataset.</p>
+    <p>App uses Microsoft Phi 3.5 Vision Model</p>
+    <p>Fine-Tuned version is built using open Burberry Product dataset.</p>
 </div>
 """
 
@@ -192,82 +187,19 @@ footer = """
 with gr.Blocks(css=CSS, theme="small_and_pretty") as demo:
     gr.HTML(TITLE)
     gr.HTML(EXPLANATION)
-    gr.DuplicateButton(value="Duplicate Space for private use", elem_classes="duplicate-button")
-    # with gr.Tab("Chatbot"):
-    #     chatbot = gr.Chatbot(height=600, placeholder=PLACEHOLDER)
-    #     gr.ChatInterface(
-    #         fn=stream_chat,
-    #         chatbot=chatbot,
-    #         fill_height=True,
-    #         additional_inputs_accordion=gr.Accordion(label="⚙️ Parameters", open=False, render=False),
-    #         additional_inputs=[
-    #             gr.Textbox(
-    #                 value="You are a helpful assistant",
-    #                 label="System Prompt",
-    #                 render=False,
-    #             ),
-    #             gr.Slider(
-    #                 minimum=0,
-    #                 maximum=1,
-    #                 step=0.1,
-    #                 value=0.8,
-    #                 label="Temperature",
-    #                 render=False,
-    #             ),
-    #             gr.Slider(
-    #                 minimum=128,
-    #                 maximum=8192,
-    #                 step=1,
-    #                 value=1024,
-    #                 label="Max new tokens",
-    #                 render=False,
-    #             ),
-    #             gr.Slider(
-    #                 minimum=0.0,
-    #                 maximum=1.0,
-    #                 step=0.1,
-    #                 value=1.0,
-    #                 label="top_p",
-    #                 render=False,
-    #             ),
-    #             gr.Slider(
-    #                 minimum=1,
-    #                 maximum=20,
-    #                 step=1,
-    #                 value=20,
-    #                 label="top_k",
-    #                 render=False,
-    #             ),
-    #             gr.Slider(
-    #                 minimum=0.0,
-    #                 maximum=2.0,
-    #                 step=0.1,
-    #                 value=1.2,
-    #                 label="Repetition penalty",
-    #                 render=False,
-    #             ),
-    #         ],
-    #         examples=[
-    #             ["How to make a self-driving car?"],
-    #             ["Give me a creative idea to establish a startup"],
-    #             ["How can I improve my programming skills?"],
-    #             ["Show me a code snippet of a website's sticky header in CSS and JavaScript."],
-    #         ],
-    #         cache_examples=False,
-    #     )
-    with gr.Tab("Vision"):
+    with gr.Tab("Burberry Vision"):
         with gr.Row():
             input_img = gr.Image(label="Input Picture")
         with gr.Row():
             model_selector = gr.Dropdown(choices=list(models.keys()), label="Model", value="justinj92/phi-35-vision-burberry")
-        with gr.Row():
-            text_input = gr.Textbox(label="Question")
+        # with gr.Row():
+        #     text_input = gr.Textbox(label="Question")
         with gr.Row():
             submit_btn = gr.Button(value="Submit")
         with gr.Row():
             output_text = gr.Textbox(label="Output Text")
 
-        submit_btn.click(stream_vision, [input_img, text_input, model_selector], [output_text])
+        submit_btn.click(stream_vision, [input_img, model_selector], [output_text])
 
     gr.HTML(footer)