Update app.py

app.py

@@ -23,26 +23,14 @@ labels = [
     "automotive", "blockchain", "biology", "chemistry", 
     "cryptocurrency", "data science", "design", "e-commerce",
     "education", "engineering", "entertainment", "environment",
-    "fashion", "finance", "food commerce", "gaming",
-    "healthcare", "history", "information technology", 
-    "legal", "machine learning", "marketing", "medicine",
-    "music", "philosophy", "physics", "politics", "real estate", "retail", 
-    "robotics", "social media", "sports", "technical",
-    "tourism", "travel"
-]
-
-tones = [
-    "formal", "positive", "negative", "poetic", "polite", "subtle", "casual", "neutral",
-    "informal", "pompous", "sustained", "rude", "sustained",
-    "sophisticated", "playful", "serious", "friendly"
-]
-
-styles = [
-    "poetry", "novel", "theater", "slang", "speech", "keywords", "html", "programming"
-]
-
-gender_number = [
-    "masculine singular", "masculine plural", "feminine singular", "feminine plural"
+    "fashion", "finance", "food commerce", "general",
+    "gaming", "healthcare", "history", "html",
+    "information technology", "IT", "keywords", "legal",
+    "literature", "machine learning", "marketing", "medicine",
+    "music", "personal development", "philosophy", "physics",
+    "politics", "poetry", "programming", "real estate", "retail", 
+    "robotics", "slang", "social media", "speech", "sports",
+    "sustained", "technical", "theater", "tourism", "travel"
 ]
 
 @lru_cache(maxsize=1)
@@ -60,7 +48,7 @@ def softmax(x):
     return exp_x / exp_x.sum()
 
 # Function to detect context
-def detect_context(input_text, threshold=0.03):
+def detect_context(input_text, threshold=0.022):
     # Encode the input text
     inputs = tokenizer([input_text], padding=True, truncation=True, return_tensors="pt")
     with torch.no_grad():
@@ -81,35 +69,48 @@ def detect_context(input_text, threshold=0.03):
 
     # If no contexts meet the threshold, default to "general"
     if not high_confidence_contexts:
-        high_confidence_contexts = [("general", 1.0)]
+        high_confidence_contexts = [("general", 1.0)]  # Assign a default score of 1.0 for "general"
 
     return high_confidence_contexts
 
-# Translation client
-translation_client = Client("Frenchizer/space_7")
-
-def translate_text(input_text):
-    # Translate the input text
-    return translation_client.predict(input_text)
+# Mock translation clients for different contexts
+def get_translation_client(context):
+    """
+    Returns the appropriate Hugging Face Space client for the given context.
+    For now, all contexts use the same mock space.
+    """
+    return Client("Frenchizer/space_7")  # Replace with actual Space paths for each context
+
+def translate_text(input_text, context):
+    """
+    Translates the input text using the appropriate model for the given context.
+    """
+    client = get_translation_client(context)
+    return client.predict(input_text)
 
 def process_request(input_text):
-    # Step 1: Translate the text
-    translation = translate_text(input_text)
-
-    # Step 2: Detect context
+    # Step 1: Detect context
     context_results = detect_context(input_text)
 
-    # Step 3: Print the list of high-confidence contexts
-    print("Contexts:", context_results)
+    # Step 2: Translate the text for each context
+    translations = {}
+    for context, score in context_results:
+        translations[context] = translate_text(input_text, context)
+
+    # Step 3: Print the list of high-confidence contexts and translations
+    print("High-confidence contexts (score >= 0.022):", context_results)
+    print("Translations:", translations)
 
-    # Return the translation and contexts
-    return translation, context_results
+    # Return the translations and contexts
+    return translations, context_results
 
 # Gradio interface
 def gradio_interface(input_text):
-    translation, contexts = process_request(input_text)
+    translations, contexts = process_request(input_text)
     # Format the output
-    output = f"Translation: {translation}\n"
+    output = "Translations:\n\n"
+    for context, translation in translations.items():
+        output += f"**{context.capitalize()}**: {translation}\n\n"
     return output.strip()
 
 # Create the Gradio interface
@@ -118,7 +119,7 @@ interface = gr.Interface(
     inputs="text",
     outputs="text",
     title="Frenchizer",
-    description="Translate text from English to French with context detection and threshold."
+    description="Translate text from English to French with context-specific models and different outputs."
 )
 
 interface.launch()