update params

app.py

@@ -21,10 +21,17 @@ url = os.getenv("AZURE_ENDPOINT")
 api_key = os.getenv("AZURE_API_KEY")
 
 
+# Default parameter values
+default_max_tokens = 4096
+default_temperature = 0.0
+default_top_p = 1.0
+default_presence_penalty = 0.0
+default_frequency_penalty = 0.0
+
 # Initialize MIME types
 mimetypes.init()
 
-def call_aml_endpoint(payload, url, api_key):
+def call_aml_endpoint(payload, url, api_key, params=None):
     """Call Azure ML endpoint with the given payload."""
     # Allow self-signed HTTPS certificates
     def allow_self_signed_https(allowed):
@@ -33,17 +40,24 @@ def call_aml_endpoint(payload, url, api_key):
 
     allow_self_signed_https(True)
     
-    # Set parameters (can be adjusted based on your needs)
-    parameters = {"temperature": 0.7}
+    # Set parameters from the UI inputs or use defaults
+    if params is None:
+        params = {
+            "max_tokens": default_max_tokens,
+            "temperature": default_temperature,
+            "top_p": default_top_p,
+            "presence_penalty": default_presence_penalty,
+            "frequency_penalty": default_frequency_penalty
+        }
     
-    # parameters = {
-    #     "max_tokens": 4096,
-    #     "temperature": 0.7,
-    #     "top_p": 0.1,
-    #     "presence_penalty": 0.0,
-    #     "frequency_penalty": 0.0,
-    #     "stream": True
-    # }
+    parameters = {
+        "max_tokens": int(params["max_tokens"]),
+        "temperature": float(params["temperature"]),
+        "top_p": float(params["top_p"]),
+        "presence_penalty": float(params["presence_penalty"]),
+        "frequency_penalty": float(params["frequency_penalty"]),
+        "stream": True
+    }
     
     if "parameters" not in payload["input_data"]:
         payload["input_data"]["parameters"] = parameters
@@ -62,6 +76,7 @@ def call_aml_endpoint(payload, url, api_key):
 
     try:
         logger.info(f"Sending request to {url}")
+        logger.info(f"Using parameters: {parameters}")
         response = urllib.request.urlopen(req)
         result = response.read().decode('utf-8')
         logger.info("Received response successfully")
@@ -72,7 +87,6 @@ def call_aml_endpoint(payload, url, api_key):
         error_message = error.read().decode("utf8", 'ignore')
         logger.error(f"Error message: {error_message}")
         return {"error": error_message}
-
 def improved_fetch_audio_from_url(url):
     """Improved function to fetch audio data from URL and convert to base64
     Args:
@@ -578,8 +592,8 @@ with gr.Blocks(theme=gr.themes.Soft(), css=css) as demo:
     description = gr.Markdown("""
     This demo allows you to interact with the [Phi-4-Multimodal AI model](https://aka.ms/phi-4-multimodal/techreport).
     You can type messages, upload images, or record audio to communicate with the AI.
-    Other demos include [Thoughts Organizer](https://microsoft-thoughtsorganizer.hf.space/), 
-    [Stories Come Alive](https://microsoft-storiescomealive.hf.space/), [Phine Speech Translator](https://microsoft-phinespeechtranslator.hf.space/), and [Phi-4-Mini Playground](https://microsoft-phi-4-mini.hf.space/)
+    Other demos include [Phi-4-Mini playground](https://huggingface.co/spaces/microsoft/phi-4-mini), [Thoughts Organizer](https://huggingface.co/spaces/microsoft/ThoughtsOrganizer), 
+    [Stories Come Alive](https://huggingface.co/spaces/microsoft/StoriesComeAlive), [Phine Speech Translator](https://huggingface.co/spaces/microsoft/PhineSpeechTranslator)
     """)
     
     # Store the conversation state with base64 data