Merge pull request #165 from davidbejarcaceres/Switch-model-unload-before-load-new-model

Clear model from memory before switch, fixes out of memory

lama_cleaner/model_manager.py

@@ -1,4 +1,5 @@
 import torch
+import gc
 
 from lama_cleaner.model.fcf import FcF
 from lama_cleaner.model.lama import LaMa
@@ -42,6 +43,12 @@ class ModelManager:
         if new_name == self.name:
             return
         try:
+            if (torch.cuda.memory_allocated() > 0):
+                # Clear current loaded model from memory
+                torch.cuda.empty_cache()
+                del self.model
+                gc.collect()
+
             self.model = self.init_model(new_name, self.device, **self.kwargs)
             self.name = new_name
         except NotImplementedError as e: