ben-louis / fisherpruning-pytorch Goto Github PK

感谢您复现这个code，我想请问下该方法是否支持二阶段的检测方法呢，比如cascade_mask_rcnn之类的。多谢！

您好，仓库里没有baseline训练文件，请问您是怎么对照剪枝效果呢？baseline是这样子训练吗？
base = [
'../base/models/resnet50.py', '../base/datasets/cifar10_bs16.py',
'../base/schedules/cifar10_bs128.py', '../base/default_runtime.py'
]

optimizer = dict(lr=0.004)
work_dir = "work_dirs/resnet50-baselilne"
load_from = "https://download.openmmlab.com/mmclassification/v0/resnet/resnet50_8xb32_in1k_20210831-ea4938fc.pth"

Firstly, thanks for your excellect job!
I had tried to prune the mobilenet, it contains many DWConv.
To support DWConv pruning， I make little changes

Otherwise, the following error will be reported

When finetuning the pruned model, there was error about the DWConv.

Hello! I saw you tested the code in torch1.8.

I tried it on RTX 3090 with torch==1.8, cudatoolkit==11.1, mmcv-full with serveral different version installed with pip, mmdet==2.17 When running on COCO in pruning stage, it always occurs

  File "tools/prune_train.py", line 195, in <module>
    main()
  File "tools/prune_train.py", line 191, in main
    meta=meta)
  File "/home/dell/programme/FisherPruning-Pytorch/mmdet/apis/train.py", line 174, in train_detector
    runner.run(data_loaders, cfg.workflow)
  File "/home/dell/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/mmcv/runner/epoch_based_runner.py", line 127, in run
    epoch_runner(data_loaders[i], **kwargs)
  File "/home/dell/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/mmcv/runner/epoch_based_runner.py", line 51, in train
    self.call_hook('after_train_iter')
  File "/home/dell/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/mmcv/runner/base_runner.py", line 307, in call_hook
    getattr(hook, fn_name)(self)
  File "/home/dell/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/mmcv/runner/hooks/optimizer.py", line 36, in after_train_iter
    runner.outputs['loss'].backward()
  File "/home/dell/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/torch/tensor.py", line 245, in backward
    torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs)
  File "/home/dell/anaconda3/envs/open-mmlab/lib/python3.7/site-packages/torch/autograd/__init__.py", line 147, in backward
    allow_unreachable=True, accumulate_grad=True)  # allow_unreachable flag
  File "/home/dell/programme/FisherPruning-Pytorch/tools/fisher_pruning.py", line 385, in compute_fisher_backward_hook
    grads = feature * grad_feature
RuntimeError: The size of tensor a (256) must match the size of tensor b (36) at non-singleton dimension 1

But when I tried on TITAN RTX with torch==1.3 as the author of this paper suggests, this error disappears.

Have you encountered this problem? Thanks!

Thank you very much for your optimization. I tried to reproduce the pruning effect on classification. But it reported an error. I suspect it is a torch version problem, but after switching to the same version as your experiment, still have this problem. Can you give suggestions?

2021-11-18 21:53:59,417 - mmcls - INFO - Environment info:

sys.platform: linux
Python: 3.7.11 (default, Jul 27 2021, 14:32:16) [GCC 7.5.0]
CUDA available: True
GPU 0,1,2,3,4,5,6,7: GeForce RTX 3090
CUDA_HOME: /usr/local/cuda-11.1
NVCC: Build cuda_11.1.TC455_06.29069683_0
GCC: gcc (GCC) 5.4.0
PyTorch: 1.8.0+cu111
PyTorch compiling details: PyTorch built with:

• GCC 7.3
• C++ Version: 201402
• Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications
• Intel(R) MKL-DNN v1.7.0 (Git Hash 7aed236906b1f7a05c0917e5257a1af05e9ff683)
• OpenMP 201511 (a.k.a. OpenMP 4.5)
• NNPACK is enabled
• CPU capability usage: AVX2
• CUDA Runtime 11.1
• NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86
• CuDNN 8.0.5
• Magma 2.5.2
• Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.1, CUDNN_VERSION=8.0.5, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.8.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON,

TorchVision: 0.9.0+cu111
OpenCV: 4.5.3
MMCV: 1.3.17
MMCV Compiler: GCC 5.4
MMCV CUDA Compiler: 11.1
MMClassification: 0.15.0+729c6c1

it only trained on Linux os？ Windows os is ok？