Update model.py

model.py

@@ -3,7 +3,10 @@ import torch
 import logging
 from typing import Dict, List
 import time
-import psutil
+try:
+    import psutil
+except ImportError:
+    psutil = None
 
 # Configure logging
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
@@ -32,9 +35,12 @@ def call_ai_model_for_insights(input_data: Dict, delay_risk: float) -> List[str]
     max_retries = 3
     retry_delay = 15  # seconds
     
-    # Log system resources
-    memory = psutil.virtual_memory()
-    logger.info(f"System memory - Total: {memory.total / 1e9:.2f} GB, Available: {memory.available / 1e9:.2f} GB, Used: {memory.percent}%")
+    # Log system resources if psutil is available
+    if psutil:
+        memory = psutil.virtual_memory()
+        logger.info(f"System memory - Total: {memory.total / 1e9:.2f} GB, Available: {memory.available / 1e9:.2f} GB, Used: {memory.percent}%")
+    else:
+        logger.warning("psutil not available; cannot log system memory usage")
     
     for attempt in range(max_retries):
         try:
@@ -67,11 +73,11 @@ def call_ai_model_for_insights(input_data: Dict, delay_risk: float) -> List[str]
             """
             
             with torch.no_grad():
-                inputs = tokenizer(prompt, return_tensors="pt", max_length=128, truncation=True).to("cpu")
+                inputs = tokenizer(prompt, return_tensors="pt", max_length=64, truncation=True).to("cpu")
                 outputs = model.generate(
                     **inputs,
-                    max_new_tokens=30,
-                    num_beams=4,
+                    max_new_tokens=20,
+                    num_beams=2,
                     temperature=0.7,
                     do_sample=True
                 )