silvandeleemput memcnn issues

silvandeleemput / memcnn

PyTorch Framework for Developing Memory Efficient Deep Invertible Networks

MIT License

252 stars 26 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Kernel crash on calling `.inverse()` on conditional normalizing flow

#79 minh-nguyenhoang opened 3 months ago
0
Where the jacobian is being computed?

#78 mjack3 opened 1 year ago
1
Slower training with trivial inverse

#77 michaelsdr opened 1 year ago
2
Fix CIFAR data loaders

#76 silvandeleemput closed 1 year ago
0
Make compatible for PyTorch 2

#75 silvandeleemput opened 1 year ago
0
AttributeError: Can't pickle local object 'get_cifar_data_loaders.<locals>.<lambda>'

#74 gbpollam closed 1 year ago
5
Error: The tensor has a non-zero number of elements, but its data is not allocated yet.

#73 llstela closed 1 year ago
2
how to joint several reversble reversible additive blocks together

#72 LoveLiveSun opened 2 years ago
1
A question on implementation

#71 jambo6 closed 2 years ago
4
ReLU before Inversible layers

#70 Wayne2Wang opened 2 years ago
1
Problems when memcnn is used with data parallel

#69 coolbay opened 3 years ago
1
Give a simple example of classification?

#68 ghost opened 3 years ago
2
64 fix affineadaptersigmoid 2dim inputs

#67 silvandeleemput closed 3 years ago
0
Cleaned tox build to fix CI

#66 silvandeleemput closed 3 years ago
0
Fixing CI

#65 silvandeleemput closed 3 years ago
0
Fixing AffineAdapterSigmoid for 2 dimensional inputs

#64 silvandeleemput closed 3 years ago
0
About AffineCoupling and AdditiveCoupling

#63 xuedue opened 3 years ago
12
What setting the gradients manually is needed?

#62 lighShanghaitech opened 3 years ago
1
Fix/update Pillow version in requirements.txt

#61 silvandeleemput closed 4 years ago
0
57 add mixed precision support

#60 silvandeleemput closed 4 years ago
0
Remove Python2, torch 0.4 and add torch1.7 CI support

#59 silvandeleemput closed 4 years ago
1
Remove Python 2.7 and add PyTorch 1.7 to CI tests

#58 silvandeleemput closed 4 years ago
0
Mixed precision and InvertibleModuleWrapper

#57 ibro45 closed 4 years ago
5
Droupout issue

#56 lighShanghaitech closed 1 year ago
4
feat(module): add reversible cropping and padding

#55 ClashLuke closed 3 years ago
0
feat(coupling): Inplace Padding

#54 ClashLuke closed 3 years ago
6
Memory demand not independent of depth because ctx.saved_tensors not freed?

#53 cetmann opened 4 years ago
2
perf: Computation profiling and optimization

#52 ClashLuke closed 3 years ago
3
Add preserve_rng_state argument to InvertibleModuleWrapper

#51 silvandeleemput closed 4 years ago
1
feat: add chunk-dim parameter

#50 ClashLuke closed 4 years ago
5
Custom split dimension

#49 ClashLuke closed 4 years ago
1
fix(revop): add support for gradless ops

#48 ClashLuke closed 4 years ago
5
About implementation_fwd and implementation_bwd

#47 djiajunustc opened 4 years ago
1
Add default preserve_rng_state=True support to InvertibleCheckpointFunction

#46 silvandeleemput closed 4 years ago
0
Multi inputs and outputs

#45 cetmann closed 4 years ago
3
Distributed Data Parallel

#44 lucidrains closed 4 years ago
11
Multi-input / multi-output reversible layers

#43 cetmann closed 4 years ago
5
Fix for gpu memory spike during backwards

#42 silvandeleemput closed 4 years ago
1
Memory demand not constant in network depth

#41 cetmann closed 4 years ago
2
Refactor revop using torch.autograd.Function instead of using hooks

#40 silvandeleemput closed 4 years ago
1
Residual connection - Trying to backward through the graph a second time

#39 ibro45 closed 4 years ago
8
Keep_input=False, keep_input_inverse=False issue both in the minimal example and a custom model

#38 ibro45 closed 4 years ago
4
Memory consumption does not seem to reflect the indicated consumption of nvidia-smi

#37 duag closed 4 years ago
9
Backing over reversible network twice

#36 lucidrains closed 4 years ago
2
Removal of Tensorboard and Tensorflow dependencies

#35 silvandeleemput closed 4 years ago
1
Update documentation for the new MemCNN API

#34 silvandeleemput closed 4 years ago
1
Major rework of the MemCNN core and API

#33 silvandeleemput closed 4 years ago
2
Bug fixes related to summary_iterator and NSamplesRandomSampler

#32 silvandeleemput closed 4 years ago
1
SummaryIterator can not be imported

#31 silvandeleemput closed 4 years ago
0
RandomSampler nsamples attribute causes error

#30 silvandeleemput closed 4 years ago
0