chendaryen / artfusion Goto Github PK

请问这样是为了保持随机性吗，谢谢

我正在使用这行命令
python main.py --name test --base ./configs/kl16_content12.yaml --basedir ./checkpoints -t True --gpus 0,1,
然后有如下报错：
(T2I) zzj@zzj:~/disk1/zzj/ArtFusion-main$ python main.py --name test --base ./configs/kl16_content12.yaml --basedir ./checkpoints -t True --gpus 0,1,
Global seed set to 2023
Running on GPUs 0,1,
DualCondLDM: Running in eps-prediction mode
DualConditionDiffusionWrapper has 181.02 M params.
Keeping EMAs of 205.
making attention of type 'vanilla' with 512 in_channels
making attention of type 'vanilla' with 512 in_channels
making attention of type 'vanilla' with 512 in_channels
Working with z of shape (1, 16, 16, 16) = 4096 dimensions.
making attention of type 'vanilla' with 512 in_channels
making attention of type 'vanilla' with 512 in_channels
making attention of type 'vanilla' with 512 in_channels
making attention of type 'vanilla' with 512 in_channels
Restored from checkpoints/vae/kl-f16.ckpt
Training DualCondLDM as an adaptive conditional model.
Keeping EMAs of 3.
GPU available: True, used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
datasets/wiki-art/train.npy
total 79432 hybrid training data, 118287 content data.
datasets/wiki-art/train.npy
total 79432 hybrid training data, 118287 content data.
accumulate_grad_batches = 8
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0,1]
Global seed set to 2023
initializing distributed: GLOBAL_RANK: 0, MEMBER: 1/2
Global seed set to 2023
initializing distributed: GLOBAL_RANK: 1, MEMBER: 2/2

distributed_backend=nccl
All distributed processes registered. Starting with 2 processes

Save project config
exception!!!!!
Summoning checkpoint.
Save last in ./checkpoints/2023-07-17T04-24-58_test/models/last.ckpt
Traceback (most recent call last):
File "main.py", line 608, in
trainer.fit(model, data)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 741, in fit
self._fit_impl, model, train_dataloaders, val_dataloaders, datamodule, ckpt_path
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 685, in _call_and_handle_interrupt
return trainer_fn(*args, **kwargs)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 777, in _fit_impl
self._run(model, ckpt_path=ckpt_path)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1199, in _run
self._dispatch()
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1279, in _dispatch
self.training_type_plugin.start_training(self)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/plugins/training_type/ddp_spawn.py", line 173, in start_training
self.spawn(self.new_process, trainer, self.mp_queue, return_result=False)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/plugins/training_type/ddp_spawn.py", line 201, in spawn
mp.spawn(self._wrapped_function, args=(function, args, kwargs, return_queue), nprocs=self.num_processes)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 240, in spawn
return start_processes(fn, args, nprocs, join, daemon, start_method='spawn')
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 198, in start_processes
while not context.join():
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 160, in join
raise ProcessRaisedException(msg, error_index, failed_process.pid)
torch.multiprocessing.spawn.ProcessRaisedException:

-- Process 0 terminated with the following error:
Traceback (most recent call last):
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 69, in _wrap
fn(i, *args)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/plugins/training_type/ddp_spawn.py", line 208, in _wrapped_function
result = function(*args, **kwargs)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/plugins/training_type/ddp_spawn.py", line 236, in new_process
results = trainer.run_stage()
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1289, in run_stage
return self._run_train()
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1306, in _run_train
self._pre_training_routine()
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1301, in _pre_training_routine
self.call_hook("on_pretrain_routine_start")
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/trainer.py", line 1495, in call_hook
callback_fx(*args, **kwargs)
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/pytorch_lightning/trainer/callback_hook.py", line 148, in on_pretrain_routine_start
callback.on_pretrain_routine_start(self, self.lightning_module)
File "/home/zzj/disk1/zzj/ArtFusion-main/main.py", line 223, in on_pretrain_routine_start
os.path.join(self.cfgdir, "{}-project.yaml".format(self.now)))
File "/home/zzj/anaconda3/envs/T2I/lib/python3.7/site-packages/omegaconf/omegaconf.py", line 223, in save
with io.open(os.path.abspath(f), "w", encoding="utf-8") as file:
FileNotFoundError: [Errno 2] No such file or directory: '/home/zzj/disk1/zzj/ArtFusion-main/checkpoints/2023-07-17T04-24-58_test/configs/2023-07-17T04-24-58-project.yaml'

在初始化第二张卡时候报错，请问能否告知解决办法。

Hello, bro, I am very interested in your research. Can the code for the first phase of VAE model training be updated simultaneously? Since the dataset I trained was not based on ImageNet or Coco, the subsequent model training results were bad, and the problem shown in the figure below occurred. Thanks for your contributions to this field. Looking forward to your reply.

In Supplementary Material, part E. Additional Qualitative Results, you mention this work can apply on higher-dimensional content images, while keeping the size of the style images at 256x256.
Can you explain or given an example on how to modift the code to generate result using 512x512 content image and 256x256 style image?
Thanks.

If I just want to test this project for training models, what should I do？