Upload inference.py

inference.py

@@ -3,7 +3,7 @@ import os
 os.environ['CUDA_LAUNCH_BLOCKING'] = "1"
 import torch
 from transformers import AutoTokenizer
-from emotional_gemma_clean import EmotionalLlamaModel, EMOTION_DIMENSIONS, EMOTION_DIMENSIONS_REFERENCE
+from emotional_gemma import EmotionalLlamaModel, EMOTION_DIMENSIONS, EMOTION_DIMENSIONS_REFERENCE
 from peft import PeftModel, PeftConfig
 
 import torch.nn.functional as F
@@ -151,7 +151,7 @@ if __name__ == "__main__":
     # --- Generation ---
     # Prepare the prompt using the chat template
     prompt = tokenizer.apply_chat_template([
-        {"role": "user", "content": "How are you feeling today?"},
+        {"role": "user", "content": "How are you?"},
     ], tokenize=False, add_generation_prompt=True)
 
     print(f"\nPrompt:\n{prompt}")