Upload folder using huggingface_hub

server/gradio_ui.py

@@ -71,42 +71,91 @@ def make_guess(guess_input: str, history: List) -> Tuple[str, str, str, str, str
         if history.empty:
             history = []
         else:
-            history = history.to_dict('records')
+            # Convert DataFrame to list of lists (not dicts!)
+            history = history.values.tolist()
 
-    history.append({
-        "Guess #": len(history) + 1,
-        "Your Guess": guess,
-        "Hint": obs.hint,
-        "Reward": f"{obs.reward:+.1f}"
-    })
+    # Append new guess as a list, not a dict
+    history.append([
+        len(history) + 1,           # Guess #
+        guess,                       # Your Guess
+        obs.hint,                    # Hint
+        f"{obs.reward:+.1f}"        # Reward
+    ])
 
     # Format status message
     if obs.hint == "correct":
         status = f"🎉 **Congratulations!** You found the number {guess} in {len(history)} attempts!"
         hint_display = "✨ **Last Hint:** 🎯 CORRECT! You won!"
+        welcome_msg = f"""
+        # 🎉 Victory!
+
+        You successfully found the secret number **{guess}**!
+
+        It took you **{len(history)} attempts**.
+
+        Click "🔄 New Game" to play again!
+        """
     elif obs.hint == "game over":
         secret = obs.metadata.get('secret_number', '?')
         status = f"😢 **Game Over!** The number was {secret}. Try again!"
         hint_display = f"❌ **Last Hint:** Out of attempts. The answer was {secret}."
+        welcome_msg = f"""
+        # 😢 Game Over
+
+        You ran out of attempts!
+
+        The secret number was **{secret}**.
+
+        Click "🔄 New Game" to try again!
+        """
     elif obs.hint.startswith("invalid"):
         status = "⚠️ **Invalid Guess!** Number must be between 1 and 100."
         hint_display = f"⚠️ **Last Hint:** {obs.hint}"
+        welcome_msg = None  # Don't update
     elif obs.hint == "higher":
         status = f"📈 The number is **HIGHER** than {guess}"
         hint_display = f"⬆️ **Last Hint:** Go HIGHER than {guess}"
+        welcome_msg = f"""
+        # 🎯 Number Guessing Game
+
+        Keep going! The number is **between {guess + 1} and 100**.
+
+        You have **{obs.attempts_remaining} attempts** remaining.
+
+        Try a higher number!
+        """
     elif obs.hint == "lower":
         status = f"📉 The number is **LOWER** than {guess}"
         hint_display = f"⬇️ **Last Hint:** Go LOWER than {guess}"
+        welcome_msg = f"""
+        # 🎯 Number Guessing Game
+
+        Keep going! The number is **between 1 and {guess - 1}**.
+
+        You have **{obs.attempts_remaining} attempts** remaining.
+
+        Try a lower number!
+        """
     else:
         status = f"🤔 Hint: {obs.hint}"
         hint_display = f"💡 **Last Hint:** {obs.hint}"
+        welcome_msg = None  # Don't update
 
     # Update other displays
     attempts = f"📊 **Attempts Remaining:** {obs.attempts_remaining}"
-    reward_display = f"🎁 **Total Reward:** {sum(float(h['Reward']) for h in history):.1f}"
 
-    # Additional context
-    welcome_msg = None  # Don't update welcome
+    # Calculate total reward safely from list of lists
+    total_reward = 0.0
+    for h in history:
+        try:
+            # h is now a list: [guess_num, guess, hint, reward_str]
+            reward_str = h[3]  # Index 3 is the reward column
+            # Remove '+' sign if present and convert to float
+            total_reward += float(reward_str.replace('+', ''))
+        except (ValueError, TypeError, IndexError):
+            pass
+
+    reward_display = f"🎁 **Total Reward:** {total_reward:.1f}"
 
     # Clear input for next guess
     next_input = ""
@@ -123,15 +172,7 @@ def create_gradio_interface() -> gr.Blocks:
             """
             # 🎯 Number Guessing Game Environment
 
-            Welcome to the **OpenEnv Number Guessing Game**! Test your reasoning skills by guessing
-            a secret number between 1 and 100. You'll get hints after each guess to help you narrow
-            down the answer.
-
-            **Perfect for:**
-            - 🤖 Training RL agents with binary search strategies
-            - 🧠 Testing reasoning and planning algorithms
-            - 📚 Learning reinforcement learning concepts
-            - 🎮 Having fun!
+            Welcome to the **OpenEnv Number Guessing Game**!
             """
         )
 

server/standalone_app.py

@@ -221,11 +221,14 @@ try:
     from gradio_ui import gradio_app
     import gradio as gr
 
-    # Mount Gradio app at /web path
-    app = gr.mount_gradio_app(app, gradio_app, path="/web")
-except ImportError:
+    # Mount Gradio app at /web path (Gradio 5.x syntax)
+    app = gr.mount_gradio_app(app, gradio_app, path="/web", root_path="/web")
+except ImportError as e:
     # Gradio not available, skip web interface
-    pass
+    print(f"Gradio not available: {e}")
+except Exception as e:
+    # Other errors during Gradio mounting
+    print(f"Error mounting Gradio: {e}")
 
 
 @app.get("/")