cleanup and some more safety

patches/external_pr/hypernetwork.py

@@ -939,10 +939,12 @@ def train_hypernetwork_tuning(id_task, hypernetwork_name, data_root, log_directo
     load_training_options = load_training_options.split(',')
     # images allows training previews to have infotext. Importing it at the top causes a circular import problem.
     for load_hypernetworks_option in load_hypernetworks_options:
+        load_hypernetworks_option = load_hypernetworks_option.strip(' ')
         if get_training_option(load_hypernetworks_option) is False:
             print(f"Cannot load from {load_hypernetworks_option}!")
             continue
         for load_training_option in load_training_options:
+            load_training_option = load_training_option.strip(' ')
             if get_training_option(load_training_option) is False:
                 print(f"Cannot load from {load_training_option}!")
                 continue

patches/external_pr/textual_inversion.py

@@ -319,10 +319,6 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st
                     del x
                     _loss_step += loss.item()
                 scaler.scale(loss).backward()
-                for group in optimizer.param_groups:
-                    for param in group["params"]:
-                        if param.grad is None:
-                            print("Found no grad!")
                 # go back until we reach gradient accumulation steps
                 if (j + 1) % gradient_step != 0:
                     continue