base_weights didn't available, it would crash when trying to use it

[And233]

both sd-script and lycoris have already been updated to the newest version

Traceback (most recent call last):
File "/home/arraneye/lora-scripts/./scripts/stable/sdxl_train_network.py", line 185, in
trainer.train(args)
File "/home/arraneye/lora-scripts/scripts/stable/train_network.py", line 262, in train
module.merge_to(text_encoder, unet, weights_sd, weight_dtype, accelerator.device if args.lowram else "cpu")
File "/home/arraneye/lora-scripts/venv/lib/python3.10/site-packages/lycoris/kohya.py", line 590, in merge_to
super().merge_to(1)
File "/home/arraneye/lora-scripts/venv/lib/python3.10/site-packages/lycoris/wrapper.py", line 569, in merge_to
lora.merge_to(weight)
File "/home/arraneye/lora-scripts/venv/lib/python3.10/site-packages/lycoris/modules/base.py", line 281, in merge_to
weight, bias = self.get_merged_weight(
File "/home/arraneye/lora-scripts/venv/lib/python3.10/site-packages/lycoris/modules/full.py", line 193, in get_merged_weight
weight, bias = self.make_weight(multiplier, device)
File "/home/arraneye/lora-scripts/venv/lib/python3.10/site-packages/lycoris/modules/full.py", line 160, in make_weight
bias = self.org_bias + diff_b * drop
TypeError: can only concatenate list (not "Tensor") to list