Comments (11)
Are you able to run testing with the provided weights? For training, 11GB memory should be enough for up to 8x upsampling, but might not for 16x.
from pacnet.
That's odd. When I run testing, it gives a very similar error.
Command:
CUDA_VISIBLE_DEVICES=4 python -m task_jointUpsampling.main --load-weights weights_flow/x8_pac_weights_epoch_5000.pth --download --factor 8 --model PacJointUpsample --dataset Sintel --data-root data/sintel
Output:
TEST LOADER START
TEST LOADER END
Model weights initialized from: weights_flow/x8_pac_weights_epoch_5000.pth
TEST START
BEFORE APPLY MODEL
BEFORE NET
AFTER NET
AFTER APPLY MODEL
BEFORE APPLY MODEL
BEFORE NET
Traceback (most recent call last):
File "/home/joseph/anaconda3/envs/pac/lib/python3.6/runpy.py", line 193, in _run_module_as_main
"__main__", mod_spec)
File "/home/joseph/anaconda3/envs/pac/lib/python3.6/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/home/joseph/pacnet-master/task_jointUpsampling/main.py", line 362, in <module>
main()
File "/home/joseph/pacnet-master/task_jointUpsampling/main.py", line 335, in main
log_test = test(model, test_loader, device, last_epoch, init_lr, args.loss, perf_measures, args) # TEST
File "/home/joseph/pacnet-master/task_jointUpsampling/main.py", line 89, in test
output = apply_model(model, lres, guide, args.factor)
File "/home/joseph/pacnet-master/task_jointUpsampling/main.py", line 23, in apply_model
out = net(lres, guide)
File "/home/joseph/anaconda3/envs/pac/lib/python3.6/site-packages/torch/nn/modules/module.py", line 493, in __call__
result = self.forward(*input, **kwargs)
File "/home/joseph/pacnet-master/task_jointUpsampling/models.py", line 245, in forward
x = self.up_convts[i](x, guide_cur)
File "/home/joseph/anaconda3/envs/pac/lib/python3.6/site-packages/torch/nn/modules/module.py", line 493, in __call__
result = self.forward(*input, **kwargs)
File "/home/joseph/pacnet-master/pac.py", line 786, in forward
self.output_padding, self.dilation, self.shared_filters, self.native_impl)
File "/home/joseph/pacnet-master/pac.py", line 498, in pacconv_transpose2d
shared_filters)
File "/home/joseph/pacnet-master/pac.py", line 252, in forward
output = torch.einsum('ijklmn,jokl->iomn', (in_mul_k, weight))
File "/home/joseph/anaconda3/envs/pac/lib/python3.6/site-packages/torch/functional.py", line 211, in einsum
return torch._C._VariableFunctions.einsum(equation, operands)
RuntimeError: CUDA out of memory. Tried to allocate 2.69 GiB (GPU 0; 10.92 GiB total capacity; 5.74 GiB already allocated; 1.86 GiB free; 2.78 GiB cached)
from pacnet.
This is indeed odd ... which pytorch version are you using? The code was originally developed for 0.4, but there is an experimental branch for 1.4 which you might try out.
from pacnet.
Using python >> import torch >> print(torch.version), mine is 1.1.0
I am running the optical flow test on Sintel data with weights_flow x8_pac_weights_epoch_50000.pth
python -m task_jointUpsampling.main --load-weights weights_flow/x8_pac_weights_epoch_5000.pth --download --factor 8 --model PacJointUpsample --dataset Sintel --data-root data/sintel
How many GB of GPU would you estimate is necessary to run the test program?
from pacnet.
11GB GPUs should be enough for both training (w/ the exception of some 16x models) and testing. versions >1.0 are not supported by the master branch (I expect some test cases to fail as well). The th14 branch is to be used with version 1.4, but has not been thoroughly tested.
from pacnet.
So, 1.0 should work then correct? Should I downgrade and test again or do you have other suggestions?
from pacnet.
You can downgrade to 1.0 or upgrade to 1.4 (and use the th14 branch).
from pacnet.
I downgraded to 1.0.0 and it still has GPU out of memory error for testing flow. Is the data-root supposed to be: --data-root data/sintel? There are a lot of folders under the data-root, should I specify a particular folder?
from pacnet.
@josephdanielchang I just tested on 11GB mem GPU and found that indeed the 8x and 16x flow tests won't work. Sorry that I didn't provide clear information before. With a 11GB mem GPU, you are able to run all depth experiments and only 4x flow experiments.
The data path is correct as is.
from pacnet.
Thanks, it does work with only 4x for flow. Followup question, where do I find the results for the these "upsampled" flow after running flow test on the sintel flow data? I only find a folder exp/sintel with test.log and train.log, but no .flo files generated anywhere. Is there supposed to be no output?
from pacnet.
Right, the code is for quantitative evaluation only and does not save results (for the semantic segmentation code though we do have a "--eval pred" option for this purpose).
from pacnet.
Related Issues (20)
- PacConv3d HOT 1
- Batch size HOT 1
- Reason for No ReLU HOT 1
- torch 1.4.0 cannot find type2backend HOT 12
- how to generate predictions using fcn8spac? HOT 10
- some basic questions HOT 3
- Questions on using the CRF layer HOT 1
- JBU training error HOT 1
- Question about Transpose
- Pytorch 1.6 autocast HOT 2
- About "Native Implemention" HOT 6
- about the crf model HOT 3
- PacConv3d and AMP HOT 2
- Reason for scaling mask input HOT 1
- PAC_CRF step setting question HOT 2
- autocast support HOT 1
- Error with pacconv: trying to differentiate twice a function that was marked with @once_differentiable
- A question about updating the weight of the kernel HOT 2
- About how to train a fcn8spac from scratch, thanks
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 pacnet.