gpleiss/efficient_densenet_pytorch

Issues

• Excuse me, what is the cause of this problem?

  #77 opened 2 years ago by SzLSzC
  1

• Question about the place of checkpoint (shared memory allocation)

  #76 opened 2 years ago by mert-kurttutan
  1

• will the inference memory reduced too?

  #75 opened 2 years ago by emergencyd
  1

• Is the normalizatin values for CIFAR-10 correct?

  #68 opened 4 years ago by abduallahmohamed
  1

• The BN running mean&var with torch.utils.checkpoint.checkpoint

  #73 opened 4 years ago by ljn114514
  2

• The function received no value for the required argument: data

  #72 opened 4 years ago by June-Luo
  2

• 网络内存消耗？

  #63 opened 4 years ago by long2double
  0

• AttributeError: module 'fire' has no attribute 'Fire'

  #71 opened 4 years ago by MRLQ-Q
  8

• How can I apply this to my own model?

  #70 opened 4 years ago by CXMANDTXW
  1

• test error interpretation

  #69 opened 4 years ago by abduallahmohamed
  1

• Is it possible to provide ImageNet pre-trained models？

  #65 opened 4 years ago by zc3945
  2

• Is this really memory efficient?

  #66 opened 4 years ago by leonardishere
  1

• pretrained densenet169 weights

  #40 opened 6 years ago by Kexiii
  8

• New adaptive pooling layer.

  #56 opened 5 years ago by CielAl
  1

• What is bn_size?

  #62 opened 5 years ago by w32zhong
  3

• dropout not in 3x3 convolutional layer

  #61 opened 5 years ago by lizhenstat
  0

• Question: why use bn_function on 1x1 conv, not on 3x3 conv

  #60 opened 5 years ago by lizhenstat
  1

• How about the version of the torchvision, project killer and pyhon-fire?

  #59 opened 5 years ago by jwnirvana
  1

• Unable to run demo.

  #58 opened 5 years ago by frost26k
  0

• Inference time issue

  #55 opened 5 years ago by youngwanLEE
  1

• Can we test using the trained model.

  #54 opened 6 years ago by TJstory
  1

• Should there be a global average pooling layer before the classifier?

  #53 opened 6 years ago by yangzy-yzy
  0

• What is the minimum GPU memory required? Still breaks for me in a single GPU

  #52 opened 6 years ago by PabloRR100
  1

• Validation dataset is being augmented as well

  #51 opened 6 years ago by garyfanhku
  1

• does this code support pytorch1.0 and the jit feature for c++ online deployment?

  #50 opened 6 years ago by mmxuan18
  3

• Could you add a License?

  #48 opened 6 years ago by csrhddlam
  1

• MultiGPU efficient densenets are slow

  #36 opened 6 years ago by wandering007
  14

• how implement memory efficient DenseNet using Tensorflow?

  #25 opened 6 years ago by phybrain
  5

• Segmentation fault (core dumped) error for multiple GPUs

  #47 opened 6 years ago by theonegis
  6

• Number of parameters doesn't match with naïve implementation

  #46 opened 6 years ago by PabloRR100
  3

• Is there any option to run ImageNet in this demo?

  #45 opened 6 years ago by jugol
  1

• torch.utils.checkpoint cost too much memory than previous 0.3 version

  #44 opened 6 years ago by mingminzhen
  1

• Softmax layer is missing in the code.

  #43 opened 6 years ago by srinidhiPY
  2

• Can't not train when using a 256*256 dataset

  #42 opened 6 years ago by nessieyang
  2

• The final test accuracy

  #34 opened 6 years ago by ustctf-zz
  6

• Multi-GPU model in pytorch0.3 consumes much more memory than pytorch0.1 version

  #31 opened 6 years ago by ZhengRui
  8

• FP become slower after upgrade to 0.4

  #38 opened 6 years ago by DesertsP
  5

• not worked in python3 environment

  #37 opened 6 years ago by tengshaofeng
  3

• Compatibility with PyTorch 0.4

  #32 opened 6 years ago by seyiqi
  11

• Test failed on PyTorch 0.3.1 with CUDA 9.0

  #33 opened 6 years ago by DesertsP
  3

• Input data size

  #30 opened 6 years ago by g5996706
  3

• Could it be MORE memory efficient?

  #22 opened 6 years ago by zhiqiangdon
  4

• I meet this problem when I run the demo.py. How to solve it?

  #24 opened 6 years ago by yyjFish
  9

• storage resize_ function

  #26 opened 6 years ago by mingminzhen
  4

• Pre-trained weight

  #27 opened 7 years ago by soon-will
  1

• Pretrained models

  #23 opened 7 years ago by EliasVansteenkiste
  2

• efficient seams not so efficient.

  #21 opened 7 years ago by tengshaofeng
  2

• Cannot reproduce the cifar100 results using models/densenet.py (not efficient)

  #16 opened 7 years ago by wandering007
  2

• Why the Error is very high?

  #20 opened 7 years ago by tryerrorman
  3

• `DenseNetEfficientMult` not giving same forwarding result as `DenseNetEfficient`

  #17 opened 7 years ago by ZhengRui
  3