Unable to do inference. Missing keys.

[RahulBhalley]

For following bash command

stylegan2_ckpt=checkpoint/stylegan2-ffhq-config-f.pt

truncation=0.8
sample=1
pics=20
alpha=0.25
seed=42
att_start=0
att_layer=18
att_channel=18

for edit in blond_hair
do
  for mask_thresh in 0.8
  do
    for train_iter in 01000 #02000 05000
    do
      python3 generate.py --ckpt $stylegan2_ckpt --truncation $truncation --sample $sample --pics $pics --alpha $alpha --seed $seed --mask_threshold $mask_thresh --train_iter $train_iter --clip_text $edit --att_layer $att_layer --att_start $att_start --att_channel $att_channel
    done
  done
done

I get missing keys errors.

size 1024 latent 512 n_mlp 8 channel_multiplier 2
Traceback (most recent call last):
  File "/kaggle/working/GanVinci/generate.py", line 192, in <module>
    feat.load_state_dict(feat_checkpoint["weights"])
  File "/opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py", line 2041, in load_state_dict
    raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for FEAT:
	Missing key(s) in state_dict: "mapper.20.weight", "mapper.20.bias", "mapper.21.weight", "mapper.21.bias", "mapper.22.weight", "mapper.22.bias", "mapper.23.weight", "mapper.23.bias", "mapper.24.weight", "mapper.24.bias", "mapper.25.weight", "mapper.25.bias", "mapper.26.weight", "mapper.26.bias", "mapper.27.weight", "mapper.27.bias", "mapper.28.weight", "mapper.28.bias", "mapper.29.weight", "mapper.29.bias", "mapper.30.weight", "mapper.30.bias", "mapper.31.weight", "mapper.31.bias", "mapper.32.weight", "mapper.32.bias", "mapper.33.weight", "mapper.33.bias", "mapper.34.weight", "mapper.34.bias", "mapper.35.weight", "mapper.35.bias".

How to set the argument values for each of the pre-trained models you've shared?
It seems attn_ prefixed arguments have something to do with these but I don't understand how.

Please help.