About the problems encountered when reproducing

[loknini]

This is great work, I ran into a problem while trying to reproduce it.
This line of code in modeling_showo in the models folder:
self.showo.resize_token_embeddings(self.vocab_size)
It keeps giving me errors, and the error message is as follows. Can anyone help me solve this problem?

`Traceback (most recent call last):
File "/mnt/d/emoemo/models/showo/inf_mmu_test.py", line 37, in
model = Showo.from_pretrained(config.model.showo.pretrained_model_path).to(device)
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/huggingface_hub/utils/_validators.py", line 114, in _inner_fn
return fn(*args, **kwargs)
File "/mnt/d/emoemo/models/showo/showo_models/modeling_utils.py", line 732, in from_pretrained
model = cls.from_config(config, **unused_kwargs)
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/diffusers/configuration_utils.py", line 260, in from_config
model = cls(**init_dict)
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/diffusers/configuration_utils.py", line 653, in inner_init
init(self, *args, **init_kwargs)
File "/mnt/d/emoemo/models/showo/showo_models/modeling_showo.py", line 41, in init
self.showo.resize_token_embeddings(self.vocab_size)
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/transformers/modeling_utils.py", line 2109, in resize_token_embeddings
model_embeds = self._resize_token_embeddings(new_num_tokens, pad_to_multiple_of, mean_resizing)
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/transformers/modeling_utils.py", line 2134, in _resize_token_embeddings
new_embeddings = self._get_resized_embeddings(
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/transformers/modeling_utils.py", line 2291, in _get_resized_embeddings
self._init_added_embeddings_weights_with_mean(
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/transformers/modeling_utils.py", line 2470, in _init_added_embeddings_weights_with_mean
is_covariance_psd = constraints.positive_definite.check(epsilon * covariance).all()
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/torch/distributions/constraints.py", line 578, in check
sym_check = super().check(value)
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/torch/distributions/constraints.py", line 555, in check
if not square_check.all():
File "/mnt/d/anaconda3/envs/emo/lib/python3.10/site-packages/torch/_meta_registrations.py", line 6471, in meta_local_scalar_dense
raise RuntimeError("Tensor.item() cannot be called on meta tensors")
RuntimeError: Tensor.item() cannot be called on meta tensors

进程已结束,退出代码1
`

[Sierkinhane]

Have you solved this problem? It's weird. I'm unsure if this problem is caused by the version conflicts of some packages.

[loknini]

Thank you for your reply, I have solved this problem. I just adjusted the version of the transformers library to match your project requirements.