Update app.py

app.py

@@ -14,16 +14,27 @@ from transformers import AutoModelForSequenceClassification, AutoTokenizer, Auto
 
 warnings.filterwarnings('ignore', category=FutureWarning)
 
+# --- Monkey Patch for Gradio Client JSON Schema Bug ---
+import gradio_client.utils as client_utils
+
+original_get_type = client_utils.get_type
+
+def patched_get_type(schema):
+    if not isinstance(schema, dict):
+        return type(schema).__name__
+    return original_get_type(schema)
+
+client_utils.get_type = patched_get_type
+# --- End of Monkey Patch ---
+
 # Download necessary NLTK data
 nltk.download('vader_lexicon', quiet=True)
 
 # ---------------------------
 # Backend Support for GGUF Models
 # ---------------------------
-# Attempt to import a llama_cpp binding if available.
-# Otherwise, fallback to the Hugging Face transformers interface.
 try:
-    # Hypothetical llama_cpp Python binding for GGUF support
+    # Attempt to import a hypothetical llama_cpp binding for GGUF models.
     from llama_cpp import Llama
     BACKEND = "llama_cpp"
 except ImportError:
@@ -34,14 +45,14 @@ except ImportError:
 # ---------------------------
 class EmotionalAnalyzer:
     def __init__(self):
-        # Load a pre-trained emotion classifier model and tokenizer
+        # Load a pre-trained emotion classifier and tokenizer
         self.emotion_model = AutoModelForSequenceClassification.from_pretrained(
             "bhadresh-savani/distilbert-base-uncased-emotion"
         )
         self.emotion_tokenizer = AutoTokenizer.from_pretrained(
             "bhadresh-savani/distilbert-base-uncased-emotion"
         )
-        # Define labels according to the model card
+        # Define the emotion labels as per the model card
         self.emotion_labels = ["sadness", "joy", "love", "anger", "fear", "surprise"]
         self.sia = SentimentIntensityAnalyzer()
 
@@ -57,7 +68,6 @@ class EmotionalAnalyzer:
         return self.sia.polarity_scores(text)
 
     def detailed_emotional_analysis(self, text):
-        """Combine VADER and TextBlob analysis for richer emotional insight."""
         vader_scores = self.sentiment_analysis(text)
         blob = TextBlob(text)
         textblob_analysis = {
@@ -74,7 +84,6 @@ class EmotionalAnalyzer:
         }
 
     def visualize_emotions(self, emotions_dict):
-        """Plot a bar chart of the current emotional state."""
         emotions_df = pd.DataFrame(list(emotions_dict.items()), columns=['Emotion', 'Percentage'])
         plt.figure(figsize=(8, 4))
         sns.barplot(x='Emotion', y='Percentage', data=emotions_df)
@@ -92,18 +101,19 @@ class LLMResponder:
     def __init__(self, model_name="SicariusSicariiStuff/Impish_LLAMA_3B_GGUF"):
         self.model_name = model_name
         if BACKEND == "llama_cpp":
-            # Initialize using llama_cpp backend (adjust parameters as needed)
-            self.llm = Llama(model_path="path/to/gguf/file.gguf", n_ctx=1024)
+            # Initialize using the llama_cpp backend.
+            # Replace "path/to/your/gguf/file.gguf" with the actual path to your GGUF file.
+            self.llm = Llama(model_path="path/to/your/gguf/file.gguf", n_ctx=1024)
             self.backend = "llama_cpp"
         else:
-            # Load using Hugging Face transformers interface
+            # Load model via Hugging Face transformers.
             self.llm_tokenizer = AutoTokenizer.from_pretrained(model_name)
             self.llm_model = AutoModelForCausalLM.from_pretrained(model_name)
             self.backend = "transformers"
 
     def generate_response(self, prompt):
         if self.backend == "llama_cpp":
-            # Use llama_cpp inference (example API call, adjust as needed)
+            # Use llama_cpp inference (adjust parameters as needed)
             result = self.llm(prompt=prompt, max_tokens=256, temperature=0.95, top_p=0.95)
             response = result.get("response", "")
         else:
@@ -124,14 +134,12 @@ class LLMResponder:
 # Main Interactive Interface Function
 # ---------------------------
 def interactive_interface(input_text):
-    # Initialize modules
     emotion_analyzer = EmotionalAnalyzer()
     llm_responder = LLMResponder()
 
-    # Perform detailed emotional analysis
+    # Perform detailed emotional analysis on the input
     emotional_data = emotion_analyzer.detailed_emotional_analysis(input_text)
-    # For demonstration, we simulate a dynamic emotional state dictionary update.
-    # In a real-world scenario, this could be updated based on conversation history.
+    # Simulate dynamic emotional state (could be updated based on conversation history)
     current_emotions = {
         'joy': random.randint(10, 30),
         'sadness': random.randint(5, 20),
@@ -142,15 +150,15 @@ def interactive_interface(input_text):
     }
     emotion_image = emotion_analyzer.visualize_emotions(current_emotions)
 
-    # Create a prompt that combines the input and the detected emotion
-    prompt = (f"Input: {input_text}\n"
-              f"Detected Emotion: {emotional_data['predicted_emotion']}\n"
-              f"VADER Scores: {emotional_data['vader']}\n"
-              "Provide a thoughtful, emotionally aware response that reflects the above data:")
-
+    # Create a prompt combining the input and detected emotion data
+    prompt = (
+        f"Input: {input_text}\n"
+        f"Detected Emotion: {emotional_data['predicted_emotion']}\n"
+        f"VADER Scores: {emotional_data['vader']}\n"
+        "Provide a thoughtful, emotionally aware response that reflects the above data:"
+    )
     llm_response = llm_responder.generate_response(prompt)
 
-    # Organize the result into a dictionary
     result = {
         'detailed_emotional_analysis': emotional_data,
         'llm_response': llm_response,
@@ -182,4 +190,3 @@ iface = gr.Interface(
 
 if __name__ == "__main__":
     iface.launch(share=True)
-