Comments (5)
After searching a lot inside the code the issue is not related with pin_memory on the dataloader but with the private method __generate_indices__ of the distributed sampler.
It is a problem even if the dataset contains 500K images.
from genforce.
Even in the case of 200K images, I get the error:
Traceback (most recent call last):
File "./train.py", line 117, in <module>
main()
File "./train.py", line 113, in main
runner.train()
File "/media/ilan/deepcomics/libs/genforce/runners/base_runner.py", line 287, in train
self.train_step(data_batch, **train_kwargs)
File "/home/libs/genforce/runners/stylegan_runner.py", line 34, in train_step
G.synthesis.lod.data.fill_(self.lod)
File "/home/miniconda3/envs/genforce/lib/python3.7/site-packages/torch/nn/modules/module.py", line 948, in __getattr__
type(self).__name__, name))
AttributeError: 'SynthesisModule' object has no attribute 'lod'
Traceback (most recent call last):
File "./train.py", line 117, in <module>
main()
File "./train.py", line 113, in main
runner.train()
File "/home/libs/genforce/runners/base_runner.py", line 287, in train
self.train_step(data_batch, **train_kwargs)
File "/home/libs/genforce/runners/stylegan_runner.py", line 34, in train_step
G.synthesis.lod.data.fill_(self.lod)
File "/home/miniconda3/envs/genforce/lib/python3.7/site-packages/torch/nn/modules/module.py", line 948, in __getattr__
type(self).__name__, name))
AttributeError: 'SynthesisModule' object has no attribute 'lod'
Traceback (most recent call last):
File "./train.py", line 117, in <module>
main()
File "./train.py", line 113, in main
runner.train()
File "/home/libs/genforce/runners/base_runner.py", line 287, in train
self.train_step(data_batch, **train_kwargs)
File "/home/libs/genforce/runners/stylegan_runner.py", line 34, in train_step
G.synthesis.lod.data.fill_(self.lod)
File "/home/miniconda3/envs/genforce/lib/python3.7/site-packages/torch/nn/modules/module.py", line 948, in __getattr__
type(self).__name__, name))
My config is the same as here: #21
from genforce.
Hi @ysig,
Thanks for your issue. Could you please set the data.repeat as a small number like 10? The variable repeats the indices of the whole dataset multiple times, requiring large memory.
from genforce.
Thank you, I will try it.
But what about the lod error: #22 (comment) ?
Thanks
from genforce.
The lod
variable is defined here. My guess is that you are using StyleGAN2 generator. Since StyleGAN2 does not require progressive training, lod
will be of no use in the generator and hence not well-defined.
from genforce.
Related Issues (20)
- Summaries in tensorboard (Image/Audio) HOT 1
- Will you provide the pretrained model of PGGAN on the FFHQ or CelebA-HQ dataset? HOT 5
- GPUs training error: Logger `logger` has already existed! HOT 9
- How we can train our own dataset using stylegan model, what should be the settings? HOT 14
- AttributeError: 'SynthesisModule' object has no attribute 'lod' HOT 9
- Training on my own data
- Training closes silently
- Randomize noise HOT 1
- about activation_scale HOT 1
- dataloader error HOT 1
- Compatibility with RTX 3060 HOT 1
- The encoder HOT 1
- How do I train using StyleGAN2? HOT 1
- Get style code (Y-space)
- StyleGAN3 support
- Resolution assert in progressive_resize_image HOT 1
- Convert class conditional models(stylegan2-ada pytorch) to genforce format? HOT 2
- Using my own dataset to retrain HOT 1
- Sudden increase in memory usage during training
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from genforce.