u1ug

Current timestamp checking verifies if block timestamp is higher than previous but it is not preventing from setting future timestamp.

File "/home/me/Paella/src/utils.py", line 77, in load_conditional_models vqgan.load_state_dict(torch.load(vqgan_path, map_location=device)['state_dict']) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^ KeyError: 'state_dict'

Got a negative loss value while was training a CLIP model: am I doing something wrong or it is a loss function bug? Here is my code ```python import torch...