Update app.py

app.py

@@ -2,8 +2,7 @@ import gradio as gr
 from PIL import Image
 from inference.main import MultiModalPhi2
 
-messages = []
-
+# Initialize the chatbot model
 multimodal_phi2 = MultiModalPhi2(
     modelname_or_path="Navyabhat/Llava-Phi2",
     temperature=0.2,
@@ -11,7 +10,38 @@ multimodal_phi2 = MultiModalPhi2(
     device="cpu",
 )
 
+# Initialize chatbot history
+messages = []
+
+# UI setup
+with gr.Blocks() as demo:
+    chatbot = gr.Chatbot(
+        [],
+        elem_id="chatbot",
+        bubble_full_width=False,
+        avatar_images=(None, None),  # You can specify avatar images if needed
+    )
+
+    with gr.Row():
+        txt = gr.Textbox(
+            scale=4,
+            show_label=False,
+            placeholder="Enter text and press enter, or upload an image",
+            container=False,
+        )
+        btn = gr.UploadButton("📁", file_types=["image", "video", "audio"])
+
+    txt_msg = txt.submit(add_content, [chatbot, txt], [chatbot, txt], queue=False).then(
+        run, [chatbot, txt, None, None, None], [chatbot, txt, None, None, None], api_name="bot_response"
+    )
+    txt_msg.then(lambda: gr.Textbox(interactive=True), None, [txt], queue=False)
+    file_msg = btn.upload(add_content, [chatbot, None, None, btn, None], [chatbot, None, None, None], queue=False).then(
+        run, [chatbot, None, None, None, btn], [chatbot, None, None, None, None]
+    )
+
+    chatbot.like(print_like_dislike, None, None)
 
+# Function to add content to chatbot
 def add_content(chatbot, text, image, audio_upload, audio_mic) -> gr.Chatbot:
     textflag, imageflag, audioflag = False, False, False
     if text not in ["", None]:
@@ -29,14 +59,10 @@ def add_content(chatbot, text, image, audio_upload, audio_mic) -> gr.Chatbot:
             audioflag = True
     if not any([textflag, imageflag, audioflag]):
         # Raise an error if neither text nor file is provided
-        raise gr.Error("Enter a valid text, image or audio")
+        raise gr.Error("Enter a valid text, image, or audio")
     return chatbot
 
-
-def clear_data():
-    return {prompt: None, image: None, audio_upload: None, audio_mic: None, chatbot: []}
-
-
+# Function to run the chatbot
 def run(history, text, image, audio_upload, audio_mic):
     if text in [None, ""]:
         text = None
@@ -55,61 +81,10 @@ def run(history, text, image, audio_upload, audio_mic):
     if image is not None:
         image = Image.open(image)
     outputs = multimodal_phi2(text, audio, image)
-    # outputs = ""
 
     history.append((None, outputs.title()))
     return history, None, None, None, None
 
-
-with gr.Blocks() as demo:
-    gr.Markdown("## MulitModal Phi2 Model Pretraining and Finetuning from Scratch")
-
-    with gr.Row():
-        with gr.Column(scale=4):
-            
-            with gr.Box():
-                with gr.Row():
-                    # Adding a Textbox with a placeholder "write prompt"
-                    prompt = gr.Textbox(
-                        placeholder="Ask anything", lines=2, label="Query", value=None
-                    )
-                
-                with gr.Row():
-                    # Adding image
-                    image = gr.Image(type="filepath", value=None)
-                
-                with gr.Row():
-                    # Add audio
-                    audio_upload = gr.Audio(source="upload", type="filepath")
-                    audio_mic = gr.Audio(
-                        source="microphone", type="filepath", format="mp3"
-                    )
-
-        with gr.Column(scale=8):
-            with gr.Box():
-                with gr.Row():
-                    chatbot = gr.Chatbot(
-                        avatar_images=("🧑", "🤖"),
-                        height=550,
-                    )
-                with gr.Row():
-                    # Adding a Button
-                    submit = gr.Button()
-                    clear = gr.Button(value="Clear")
-
-    submit.click(
-        add_content,
-        inputs=[chatbot, prompt, image, audio_upload, audio_mic],
-        outputs=[chatbot],
-    ).success(
-        run,
-        inputs=[chatbot, prompt, image, audio_upload, audio_mic],
-        outputs=[chatbot, prompt, image, audio_upload, audio_mic],
-    )
-
-    clear.click(
-        clear_data,
-        outputs=[prompt, image, audio_upload, audio_mic, chatbot],
-    )
-
+# Launch the Gradio UI
+demo.queue()
 demo.launch()