Update app.py

app.py

@@ -1,5 +1,6 @@
 import gradio as gr
 from transformers import pipeline
+import os
 
 # --- Model Loading ---
 # Using the model ID you've been working with, which is fine-tuned on an English dataset (IMDB).
@@ -8,8 +9,9 @@ MODEL_ID = "Light-Dav/sentiment-analysis-full-project"
 
 try:
     # Attempt to load the pipeline. This needs to be outside the function for efficiency.
-    # The 'return_all_scores=True' is important for getting scores for all labels.
-    sentiment_analyzer = pipeline("sentiment-analysis", model=MODEL_ID, return_all_scores=True)
+    # Using top_k=None to get all scores (equivalent to return_all_scores=True in older versions).
+    # This also addresses the UserWarning about `return_all_scores` deprecation.
+    sentiment_analyzer = pipeline("sentiment-analysis", model=MODEL_ID, top_k=None)
     model_loaded_successfully = True
 except Exception as e:
     print(f"Error loading model: {e}")
@@ -81,21 +83,24 @@ def interpret_sentiment(label, score):
     description = ""
     color_class = ""
 
-    if label.lower() == "positive": # Your model might output 'LABEL_2' or 'POS'
+    # IMPORTANT: Adjust 'LABEL_0', 'LABEL_1', 'LABEL_2' to your model's actual output labels
+    # Check your model's config.json on Hugging Face Hub under 'id2label' or 'label2id'
+    # Example: "id2label": {"0": "negative", "1": "neutral", "2": "positive"}
+    if label.lower() == "positive" or label.lower() == "label_2": # Assuming LABEL_2 is positive
         emoji = "😊"
         description = "This text expresses a **highly positive** sentiment." if score > 0.9 else "This text expresses a **positive** sentiment."
         color_class = "sentiment-positive"
-    elif label.lower() == "negative": # Your model might output 'LABEL_0' or 'NEG'
+    elif label.lower() == "negative" or label.lower() == "label_0": # Assuming LABEL_0 is negative
         emoji = "😠"
         description = "This text expresses a **highly negative** sentiment." if score > 0.9 else "This text expresses a **negative** sentiment."
         color_class = "sentiment-negative"
-    elif label.lower() == "neutral": # Your model might output 'LABEL_1' or 'NEUTRAL'
+    elif label.lower() == "neutral" or label.lower() == "label_1": # Assuming LABEL_1 is neutral
         emoji = "😐"
         description = "This text expresses a **neutral** sentiment."
         color_class = "sentiment-neutral"
     else:
         emoji = "❓"
-        description = "Could not confidently determine sentiment."
+        description = "Could not confidently determine sentiment. Unexpected label."
         color_class = ""
     
     return f"<div class='sentiment-display {color_class}'>{emoji} {label.upper()} ({score:.2f})</div>" + \
@@ -118,8 +123,8 @@ def analyze_sentiment(text):
         }
 
     try:
-        # The pipeline returns a list of dictionaries if return_all_scores=True
-        # e.g., [[{'label': 'LABEL_0', 'score': 0.9}, {'label': 'LABEL_1', 'score': 0.05}]]
+        # The pipeline returns a list of lists of dictionaries if top_k=None
+        # e.g., [[{'label': 'LABEL_0', 'score': 0.9}, {'label': 'LABEL_1', 'score': 0.05}, ...]]
         results = sentiment_analyzer(text)[0] # Get the first (and only) list of results
 
         # Sort results by score in descending order
@@ -178,26 +183,34 @@ with gr.Blocks(css=custom_css, theme=gr.themes.Soft()) as demo: # Using gr.Block
                     ["The meeting concluded without any major decisions."]
                 ],
                 inputs=text_input,
+                fn=analyze_sentiment, # <-- CORRECTED: Function to run for examples
+                outputs=[overall_sentiment_output, confidence_scores_output, raw_output], # <-- CORRECTED: Outputs for examples
                 cache_examples=True # Caching examples for faster loading
             )
 
         with gr.Column(scale=3):
             gr.Markdown("## 📈 Analysis Results 📉")
-            overall_sentiment_output = gr.HTML(label="Overall Sentiment") # Using HTML to render custom styled sentiment
-            confidence_scores_output = gr.Label(num_top_classes=3, label="Confidence Scores") # Default Gradio Label
-            raw_output = gr.JSON(label="Raw Model Output", visible=False) # For debugging/advanced users
+            # Using gr.HTML for "Overall Sentiment" to allow custom CSS and rich content
+            overall_sentiment_output = gr.HTML(label="Overall Sentiment")
+            # Using gr.Label for "Confidence Scores" as it's designed for classification outputs
+            confidence_scores_output = gr.Label(num_top_classes=3, label="Confidence Scores")
+            # Using gr.JSON for raw output, useful for debugging, but hidden by default
+            raw_output = gr.JSON(label="Raw Model Output", visible=False)
 
     # --- Event Listeners ---
+    # Trigger analysis when text input changes (optional, can be resource intensive if live=True)
     text_input.change(
         fn=analyze_sentiment,
         inputs=text_input,
         outputs=[overall_sentiment_output, confidence_scores_output, raw_output],
         # live=True # Uncomment for live updates as user types (can be resource intensive)
     )
+    # Trigger analysis when the button is clicked
     analyze_btn.click(
         fn=analyze_sentiment,
         inputs=text_input,
         outputs=[overall_sentiment_output, confidence_scores_output, raw_output]
     )
 
+# Launch the Gradio application
 demo.launch()