ssnl / dataset-distillation

Open-source code for paper "Dataset Distillation"

https://ssnl.github.io/dataset_distillation

ssnl/dataset-distillation Issues

Possible bugs
Closed a year ago8
Question about the updata of the synthetic data
Closed a year ago2
RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 0. Got 4035 and 4003 in dimension 2 at /opt/conda/conda-bld/pytorch_1544199946412/work/aten/src/TH/generic/THTensorMoreMath.cpp:1333
Closed 2 years ago2
RuntimeError: CUDA error: device-side assert triggered
Closed 2 years ago1
How the custom dataset be structured ?
Closed 2 years ago1
AttributeError: module 'torch._C' has no attribute '_cuda_setDevice'
Closed 2 years ago2
How to apply it on the custom dataset?
Closed 2 years ago1
RuntimeError: CUDA out of memory.
Closed 2 years ago7
The size of tensor a (64) must match the size of tensor b (32) at non-singleton dimension 0
Closed 2 years ago5
Question about distill my own dataset
Closed 2 years ago2
How to use it for custom model and dataset?
Closed 3 years ago2
Broken: yaml.load(input) is removed in PyYAML >=6.0
Closed 2 years ago
Question about weights update in network of phase test
Updated 2 years ago4
Question about commands
Closed 2 years ago5
Question about distilled images
Closed 2 years ago2
Adapt distill on dataset SVHN
Closed 2 years ago2
Is the training process on distilled data can be conducted in main.py
Closed 2 years ago3
How to distribute different GPUs for some large models
Updated 2 years ago1
BatchNorm2d of ResNet model warning
Closed 2 years ago5
Question about pulling data from a .gz file
Closed 2 years ago2
Uniquenamespace
Closed 2 years ago1
Running in Jupyter
Closed 3 years ago1
Applicability to general structured datasets
Updated 3 years ago
retrain distilled images with minibatch-SGD
Updated 3 years ago4
Question about applicability
Updated 4 years ago
Good luck in next sumission
Updated 4 years ago3
Loss becomes NaN
Closed 5 years ago5
How to adapt to our own databases/architecture?
Closed 4 years ago8
How to get the distilled images?
Closed 4 years ago
back-gradient optimization technique
Closed 4 years ago1
Cannot find definition of `local_n_nets`
Closed 4 years ago2
Max number of classes
Closed 4 years ago2
How to use a custom dataset?
Closed 4 years ago1
cifar10: RuntimeError: Mismatch in shape
Closed 4 years ago3
Logger warnings and order of optimizer.step() and lr_scheduler.step()
Closed 5 years ago3
Compare to training on randomly selected samples
Closed 5 years ago2
How to distill dataset of size exceeding gpu memory size limit
Closed 5 years ago4
Distilling labels as well as images
Closed 5 years ago4
Structure of results.pth
Closed 5 years ago2
Getting distilled images and testing on them
Closed 5 years ago
Questions about training after dataset distillation
Closed 5 years ago1
How do you keep buffer fixed during gradient steps
Closed 5 years ago1
Questions about implementation of optimizing distilled data
Closed 5 years ago3
Does the instillation model and the test model must be the same?
Closed 5 years ago2
Getting what appears to be noise on Imagenette + XResnet
Closed 5 years ago8
Where do distilled images get saved?
Closed 5 years ago2
some question about dataset distillation
Closed 5 years ago2
ImportError: cannot import name 'reduce_op'
Closed 5 years ago4
Replicating results
Closed 5 years ago3
Bug in kmeans baseline function?
Closed 5 years ago2