import json from huggingface_hub import AsyncInferenceClient from core import config from core.globals import ml_models async def predict_sentiment(text: str): client: AsyncInferenceClient = ml_models.get("llm_client") if not client: raise RuntimeError("LLM Client not loaded.") system_prompt = f"You are a strict classifier. Classify the user's text into one of these sentiments: {config.SENTIMENT_CLASSES}. Return ONLY a JSON object with 'prediction' (string) and 'confidence' (float between 0.0 and 1.0). Do not include any other text." user_prompt = text try: response = await client.chat_completion( messages=[{"role": "system", "content": system_prompt}, {"role": "user", "content": user_prompt}], max_tokens=50, temperature=0.1 ) content = response.choices[0].message.content.strip() # Clean potential markdown wrapping if content.startswith("```json"): content = content[7:] if content.endswith("```"): content = content[:-3] data = json.loads(content) return data.get("prediction", "Neutral"), float(data.get("confidence", 0.0)) except Exception as e: print(f"Sentiment classification failed: {e}") return "Neutral", 0.0 async def predict_ticket(text: str): client: AsyncInferenceClient = ml_models.get("llm_client") if not client: raise RuntimeError("LLM Client not loaded.") system_prompt = f"You are a strict classifier. Classify the user's text into one of these ticket categories: {config.TICKET_CLASSES}. Return ONLY a JSON object with 'prediction' (string) and 'confidence' (float between 0.0 and 1.0). Do not include any other text." user_prompt = text try: response = await client.chat_completion( messages=[{"role": "system", "content": system_prompt}, {"role": "user", "content": user_prompt}], max_tokens=50, temperature=0.1 ) content = response.choices[0].message.content.strip() if content.startswith("```json"): content = content[7:] if content.endswith("```"): content = content[:-3] data = json.loads(content) return data.get("prediction", "General Inquiry"), float(data.get("confidence", 0.0)) except Exception as e: print(f"Ticket classification failed: {e}") return "General Inquiry", 0.0