train_sampler.py - IndexError

[tlsdmswn01]

Hello!

I'm writing because I've been encountering an unknown error in the "train_sampler.py" script. Currently, I'm running the following code:

bash
Copy code
!python train_sampler.py simpsons_face --iter 100 --exstyle_path /home/alpaco/sej/DualStyleGAN/checkpoint/simpsons_face/refined_exstyle_code.npy
During execution, I get the following error:

plaintext
Copy code
Load options
exstyle_path: /home/alpaco/sej/DualStyleGAN/checkpoint/simpsons_face/refined_exstyle_code.npy
iter: 500000
model_name: sampler.pt
model_path: ./checkpoint/
style: simpsons_face

---

Load extrinsic style codes successfully!
Traceback (most recent call last):
File "train_sampler.py", line 56, in
icptc.train_icp(int(500000/WC.shape[0]))
File "/home/alpaco/sej/DualStyleGAN/model/sampler/icp.py", line 118, in train_icp
self.icp.train(self.f_np, uncca_opt_params)
File "/home/alpaco/sej/DualStyleGAN/model/sampler/icp.py", line 59, in train
self.train_epoch(z_np, epoch)
File "/home/alpaco/sej/DualStyleGAN/model/sampler/icp.py", line 99, in train_epoch
e = torch.from_numpy(e_np[indices[rp[idx_np]]]).float().cuda()
IndexError: index -4709754648454915060 is out of bounds for axis 0 with size 1792
I keep encountering an IndexError, and I'm not sure how to resolve it. Any suggestions?

If you could provide an advanced English translation of this content, that would be great.