get rid of del for models

handler.py

@@ -261,13 +261,13 @@ class EndpointHandler():
         total_mem = mem_info.total / (1024 ** 3)  # Convert to gigabytes
         logger.info(f"Used Memory after model 1 loaded: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
         logger.info(f"CPU Usage after model 1 loaded: {cpu_percent}%")
-        del uptake_model
-        cpu_percent = psutil.cpu_percent()
-        mem_info = psutil.virtual_memory()
-        used_mem = mem_info.used / (1024 ** 3)  # Convert to gigabytes
-        total_mem = mem_info.total / (1024 ** 3)  # Convert to gigabytes
-        logger.info(f"Used Memory after model 1 deleted: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
-        logger.info(f"CPU Usage after model 1 deleted: {cpu_percent}%")
+        # del uptake_model
+        # cpu_percent = psutil.cpu_percent()
+        # mem_info = psutil.virtual_memory()
+        # used_mem = mem_info.used / (1024 ** 3)  # Convert to gigabytes
+        # total_mem = mem_info.total / (1024 ** 3)  # Convert to gigabytes
+        # logger.info(f"Used Memory after model 1 deleted: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
+        # logger.info(f"CPU Usage after model 1 deleted: {cpu_percent}%")
         # Reasoning
         reasoning_model = ReasoningModel(
             self.device, self.tokenizer, self.input_builder)
@@ -279,13 +279,13 @@ class EndpointHandler():
         logger.info(f"Used Memory after model 2 loaded: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
         logger.info(f"CPU Usage after model 2 loaded: {cpu_percent}%")
         # print(f"CPU Usage after model 2 loaded: {cpu_percent}%")
-        del reasoning_model
+        # del reasoning_model
         cpu_percent = psutil.cpu_percent()
-        mem_info = psutil.virtual_memory()
-        used_mem = mem_info.used / (1024 ** 3)  # Convert to gigabytes
-        total_mem = mem_info.total / (1024 ** 3)  # Convert to gigabytes
-        logger.info(f"Used Memory after model 2 deleted: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
-        logger.info(f"CPU Usage after model 2 deleted: {cpu_percent}%")
+        # mem_info = psutil.virtual_memory()
+        # used_mem = mem_info.used / (1024 ** 3)  # Convert to gigabytes
+        # total_mem = mem_info.total / (1024 ** 3)  # Convert to gigabytes
+        # logger.info(f"Used Memory after model 2 deleted: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
+        # logger.info(f"CPU Usage after model 2 deleted: {cpu_percent}%")
         # print(f"CPU Usage after model 2 deleted: {cpu_percent}%")
         # Question
         question_model = QuestionModel(
@@ -298,12 +298,12 @@ class EndpointHandler():
         total_mem = mem_info.total / (1024 ** 3)  # Convert to gigabytes
         logger.info(f"Used Memory after model 3 loaded: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
         # print(f"CPU Usage after model 3 loaded: {cpu_percent}%")
-        del question_model
+        # del question_model
         cpu_percent = psutil.cpu_percent()
-        logger.info(f"CPU Usage after model 3 deleted: {cpu_percent}%")
-        mem_info = psutil.virtual_memory()
-        used_mem = mem_info.used / (1024 ** 3)  # Convert to gigabytes
-        total_mem = mem_info.total / (1024 ** 3)  # Convert to gigabytes
-        logger.info(f"Used Memory after model 3 deleted: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
+        # logger.info(f"CPU Usage after model 3 deleted: {cpu_percent}%")
+        # mem_info = psutil.virtual_memory()
+        # used_mem = mem_info.used / (1024 ** 3)  # Convert to gigabytes
+        # total_mem = mem_info.total / (1024 ** 3)  # Convert to gigabytes
+        # logger.info(f"Used Memory after model 3 deleted: {used_mem:.2f} GB, Total Mem: {total_mem:.2f} GB")
         # print(f"CPU Usage after model 3 deleted: {cpu_percent}%")
         return transcript.to_dict()